Subscribe
Share
Share
Embed
TBPN is a live tech talk show hosted by John Coogan and Jordi Hays, streaming weekdays from 11–2 PT on X and YouTube, with full episodes posted to Spotify immediately after airing.
Described by The New York Times as “Silicon Valley’s newest obsession,” TBPN has interviewed Mark Zuckerberg, Sam Altman, Mark Cuban, and Satya Nadella. Diet TBPN delivers the best moments from each episode in under 30 minutes.
You're welcome to TBPN.
Speaker 2:Today is Monday, March 9. We are live from the TBPN UltraDome. Temple of Technology, Fortress of Finance, ramp.com, Time to Money, Safe Rugby, Corporate Cards, Bill Pay, Accounting, and a whole lot more all in one place. We have a fantastic show. We were able to get Alex Epstein to come down to the Ultradome in person on short notice to get us up to speed on what's happening in the oil markets.
Speaker 2:Of course, oil is all it's it's gushing all over the timeline, all over the Financial Times, all over The Wall Street Journal as part of the
Speaker 1:And Alex had a feeling that oil would be important Yes. Future Yes. When he wrote
Speaker 3:the book Fossil Future.
Speaker 2:Yes. Which we have right here, which if you haven't given it a read, go pick a copy up, and stay tuned for him joining us at 01:10 today. Let's pull up the linear lineup and tell you who we have coming on the show today. Linear, of course, is the system for modern software development. 70% of enterprise workspaces on Linear are using agents now.
Speaker 1:We have doctor Alex Wissner-Gross over on Eoghan Systems. Yeah. He went very viral over the weekend for connecting a biologically derived fruit fly connectum Yeah. To a Mujuko physics simulated body. And we're gonna be talking all about simulating fruit flies Yes.
Speaker 1:And what will come
Speaker 2:after We're also going over Microsoft Copilot, Cowork, Julian
Speaker 1:Which we're gonna call Coco. Coco.
Speaker 2:Okay. Disney might like a word. They have IP in that space in particular, but we'll see.
Speaker 1:But do they have it in the enterprise?
Speaker 2:I don't think so. Julien is joining from Sequoia Capital. Eoghan McCabe from from Intercom, our partner, is joining because they just raised a massive
Speaker 1:cleaned 250.
Speaker 2:Is launching PAX VC. She's she left Andreessen Horowitz, I think, about a year ago and has been is now launching a new early stage fund. Anyway, let's go through the timeline and see how people are processing the news of the Strait Of Hormuz and the oil price spikes.
Speaker 3:Fox
Speaker 1:News this morning. Yes. They were saying just basically their advice to the to the captains was to man up and just send it. Yeah. Send it.
Speaker 2:That was Send it.
Speaker 1:Seems Please
Speaker 2:stay safe.
Speaker 1:I'll read the actual quote. A guy Brian said, if you wanna diminish the Iranian threat, if you wanna make sure this ends with complete capitulation, show some guts and go through that straight.
Speaker 2:That seems very, very risky right now. Stay safe out there if you are in a shipping vessel. Do whatever is responsible. There are some crazy twists happening. So apparently ships in The Gulf are declaring themselves as Chinese vessels to dodge attack.
Speaker 2:This is from the financial the Financial Times. At least 10 vessels have changed transponder messages in an apparent attempt to avoid becoming targets. A clutch of vessels trapped in The Gulf under enemy fire are adopting a tried and tested ruse to avoid attacks. They're changing flags, using transponders to declare themselves to be Chinese. There's there's always been a very odd, like, just like tug of war between how ships identify themselves because often, for tax reasons, they're bought in one country, operated by individuals from another company, but they fly a different flag to be able to go from one place to another, and it's all based on, like, the port systems.
Speaker 2:I don't fully understand it, but it is very interesting. At least 10 ships over the past week have altered their destination signal to read Chinese owner, all Chinese crew, or Chinese crew on board. About 1,000 ships are currently shut inside the Gulf and its immediate surroundings with a cumulative value of $25,000,000,000.
Speaker 1:And I don't know I don't know if you've seen some of the the maps that show I have. The strait
Speaker 2:No.
Speaker 1:Where it it looks like nothing is actually moving through. Yeah. I think in actuality, a number of the ships are actually turning off their transponders. Oh. So you can't see the movement.
Speaker 1:Yeah. Because they're basically moving a little bit Mhmm. Going through the strait Mhmm. And then turning back on again.
Speaker 2:Well, there are there are plenty of outlets that are covering the the Iran conflict in detail. We, of course, will be focused on the the business and technology impacts. I wrote about what oil what what oil prices mean for the AI build out and data centers. It's just sort of interesting to dig into the deeper supply chain of artificial intelligence. But there are some posts that we should go through around the oil story.
Speaker 2:So crude oil is five standard deviations above its fifty day moving average. Statistically speaking, this occurs every nine thousand five hundred years. So the last time would have been about six thousand years before Moses parted the Red Sea. Imagine what that did to shipping in the area. Fanciful.
Speaker 2:What else is going on? Kobe SE. The Kobe SE letter has been posting a lot about oil prices. US oil prices are now expected to rise above $100 a barrel this month. That already happened.
Speaker 2:This was posted on March 6 with markets pricing a 66% chance that it happens. The last time oil prices were above 100 was July 2022.
Speaker 1:Let's pull up this clip from Landman.
Speaker 2:I haven't seen Landman. Have you watched it? Is it good? Let's
Speaker 4:Well, you want oil to live above 60, but below 90. Mhmm. And don't get me wrong. We're still printing money at 90, but gas gets up over $3.50 a gallon, it starts to pinch. It hits a 100.
Speaker 4:Every product in America has to readjust its price. $78 a barrel. That's about perfect. You know, it brings enough profit to keep exploring, but it don't sting as much at the pump. Less, of course, you're in California.
Speaker 4:I mean, they tax the shit out of it out there. It could be $45 a barrel, and it's still $4 a pump. I don't know how those son bitches do it out there.
Speaker 2:It's movie day. 2020 TBPN again.
Speaker 4:A barrel of oil was worthless, and this place became a ghost town. And nobody's immune. Kids have to quit college. Trucks get sold or repoed.
Speaker 1:We can pause it after. We got we got the gist of it.
Speaker 2:Yeah. Well, this is the largest How
Speaker 1:how old were you when you discovered that gas is really expensive in California, not just because there's a lot of demand for it?
Speaker 2:I mean, I discovered it when I was filling up in Montana, and it was, like, $2 a gallon, and it's, like, $5 a gallon here. That was pretty wild. The the the first major gas price shock that I noticed was hurricane Katrina in my life. I think it was in high school. Because I was too young
Speaker 1:I was talking about I was talking about just the fact realizing that Gas is so expensive Yeah. Because of the taxes that California puts on it.
Speaker 2:Yeah. Because it's nice to drive around here. Put the top down, the weather's good. So they're like, yeah, you're gonna pay. You're gonna pay more for the joys of driving an internal combustion engine car.
Speaker 2:Now, the price in California is is aggressive. I don't know the structure of the prices, though. Is it percentage based or fixed? Because that affects how much the price will move based on oil price shocks. So because if the price per barrel doubles, but the but the tax is flat, you don't feel it as much here as you do other places.
Speaker 2:So I don't know. Anyway
Speaker 1:I'm pulling it up, but you can run through this.
Speaker 2:Largest supply shock by a factor of four. So the Hormuz blockade, which is current, 20,000,000 barrels were lost in supply. The Iranian revolution in 1978 was 5,600,000. The Yom Kippur War embargoes in '73 was 4.4. The Iraq Kuwait war was 4.3 in 1990.
Speaker 2:The Iran Iraq war in 1980 was four point o. And then the Ukraine Russia invasion in 2022, which is the last time that oil spiked over a $100 a barrel, was one to three. So an absolutely huge supply shock, and I'm sure it will have a lot of implications all over the economy. With triple digit prices, here's what's going to happen now, says PolicyTensor. Markets will scream when they open tomorrow.
Speaker 2:VIX will surge to levels beyond what we saw in April. The sell off will continue for some time as intermediaries shed risk. And the markets are red. They have been screaming today. The VIX futures curve already has inverted, bid up by dealers looking for insurance.
Speaker 2:This predicts a massive sell off. The pressure on this captured White House now pressure on this captured White House now beings in earnest that's kind of sort of oddly written anyone can tell them that if this persists for very long, it will destroy the Trump presidency and gut the GOP for a generation. The controlling factor here, as I have told you over and over again, is that The United States does not have the military means to reopen Hormuz. There's no military solution in sight. This means that not only does Iran have the strategic upper hand now, it means that Iran enjoys the unambiguous strategic advantage.
Speaker 2:All they need to do is keep the thing closed until he capitulates. I put my neck out far call this in advance. And someone told him yesterday I was in the minority, perhaps even a minority of one. No longer. I was correct.
Speaker 2:Just calling a shot. The blob heads and scribes were incorrect in their assessment of the strategic situation, and now markets will price to reflect reality. And we were we briefly touched on Scott Sumner's blog post on Substack at the end of Thursday's show, maybe Friday's show. We didn't get a chance to read it. I actually read through it earlier today, and it's pretty interesting.
Speaker 2:It it basically makes the case you know, it has a very controversial tendency to freak out, and and and the and the thought is that, like, he's doom posting. Maybe it's about AI. Maybe it's about this particular conflict. He is more just reflecting on this dynamic between when the market freaks out, it acts as a moderating effect to policy. And so he gives a number of examples around, like, tariffs.
Speaker 1:Or more specifically, the admin.
Speaker 2:Yeah. Yeah. Like like like the tariffs caused this massive circuit breaker, five percent sell off in the market, and then that was internalized and and very quickly adjusted. And there were a whole bunch of different carve outs to sort of, like, create soft landings. And so he's actually and what what he's getting at is that after the fact, a lot of people say, look.
Speaker 2:You didn't need to freak out because the taco happened. Trump always chickens out. The the actual proposed policy effect didn't go into didn't go into effect. And his point is that, well, it's precisely because people freaked out that it didn't go into effect. So freaking out is good in Scott Sumner's mind, at least.
Speaker 2:Anyway, I love this.
Speaker 1:Yeah. We have a solution.
Speaker 2:We have a solution. If if you're feeling the pain at the pump, pivot and get a horse. At oil at one ten, the urban horse is the only option. Pulling up to the gas station on a horse is is truly elite. I I do wanna know, what's the TCO on a horse with the food and the stables versus, you know, just keeping a what is that?
Speaker 2:A Ford Taurus in your I I don't know. That's something else. Yeah. In your garage. The horse really mogs at the at the gas station in particular.
Speaker 2:Right? Because it's just making everyone feel so stupid. Yeah. We gotta go back. One horsepower is all you need.
Speaker 2:Nick Carter says, seeing a lot of nonprocessed trusting panic ins on the TL. And and then he followed up by saying, that's what I thought because the oil prices spiked up, and then they fell down. And we did not get a $120 a barrel oil. We got exactly 100, which was still
Speaker 1:It's not like it's way, way, way too early to celebrate or anything.
Speaker 2:Yeah. Exactly. There's an interesting chart here, a really shocking chart from Oliver Grobe, unreal numbers, about Germany's nuclear power generation. So if you scroll down, look at this curve. This is truly the bell curve meme or something like that.
Speaker 2:JPMorgan estimates that had Germany not phased out nuclear power, the country would have generated 50% less electricity from fossil fuels and 84% less electricity from natural gas in 2024. Electricity prices in Germany would have been around 25% lower, and the country would have imported half as much electricity. And just complete rise and fall of nuclear power generation in Germany. One of the craziest graphs. You don't see graphs like that very often in new technologies.
Speaker 2:Typically, you see s curves or you see exponentials. No one considers the the models get better and then they get dumber. That's certainly the funniest outcome.
Speaker 1:Yeah.
Speaker 2:Very, very rough. Goldman Sachs sent a note to investors saying if oil prices increase by $10 and remain elevated for three months, US year over year headline CPI inflation would likely rise from 2.4% in January to 3% in May. Those are small numbers, but we're looking at an oil price increase of maybe $30.40 dollars over baseline, potentially higher. We don't know where oil is going to land. And so you have this weird tug of war right now with the Fed where if oil prices go up, inflation goes up, the Fed has a mandate to curb inflation.
Speaker 2:That means higher interest rates. At a time when the labor market is shedding jobs, you would expect a Fed rate cut or a lot of people are optimistic about a rate cut. Trump certainly wants rate cuts. But if inflation is climbing, there's really no solution other than keeping rates high or even raising them further. So a real jam in terms of federal monetary policy, Fed monetary policy.
Speaker 2:Here's Art Cashin at the opening bell. This is a historical video. When was this? This was a long time ago. Let's play this clip.
Speaker 4:This may be it. First, let me start out. Moratorite Salutamis S. A. And you know that's the gladiator salute.
Speaker 4:We who are about to die salute you. So it's gonna be a tough morning. This may be it.
Speaker 2:We were about to
Speaker 1:die. Salute you.
Speaker 2:Insane insane aura for you know, CNBC really doesn't get enough credit for being so innovative in terms of broadcasting and entertainment. Really, some of the some of the greatest clips
Speaker 1:I love that. One of the one of the top comments. This is an old clip. Yeah, obviously. Sleep brother.
Speaker 2:It's probably from the nineties. So Zero Hedge says, know, with oil at a 111 total panic and memetic Sisyphus shares a clip that says, half of this site for the last week has sounded like this. Let's
Speaker 5:Producers are telling me there is breaking news. The Asian financial markets have just opened to a huge sell off, and we're gonna switch to that story right now.
Speaker 2:Good.
Speaker 3:I'm glad I'm here.
Speaker 5:Your thoughts, Tracy Jordan, on how this is gonna impact Wall Street.
Speaker 3:Larry, I'm not an expert, but I do have a strong opinion. New York, as we know it, will no longer exist tomorrow.
Speaker 2:Is this from Thirty Rock? That's so good.
Speaker 1:That's so
Speaker 2:funny. Oil came way came way in from its overnight highs, says Joe Wissenthal. The quote post is Chris Paul hits
Speaker 1:a huge three to cut down the lead to 42.
Speaker 2:Absolutely. Absolutely brutal. So let me tell you about Gusto, the unified platform for payroll benefits and HR built to evolve with modern with modern small and medium sized businesses. Stop making fun
Speaker 1:of We're we're somewhat of a smodern business. Modern, medium. And modern.
Speaker 2:Small and modern. It's modern. Let me also tell you about CrowdStrike. Your business is AI. Their business is securing it.
Speaker 2:CrowdStrike secures AI and stops breaches. So
Speaker 1:We gotta give you a shout out to George Kurtz for the Mercedes performance yesterday.
Speaker 2:Oh, yeah.
Speaker 1:One two. One two.
Speaker 2:Really good. George Kurtz on
Speaker 1:Would you look at that? Takes the stake in the team. Suddenly, they're at the top.
Speaker 2:Yeah. He he's on to something. Well, the my essay this morning was titled, why is no one talking about oil? Of course, everyone is talking about oil. Oil discourse gushed onto the timeline this weekend.
Speaker 2:Crude prices spiked to nearly a $120 a barrel as a broadening war in Iran threatens both transportation routes and production. The geopolitical and economic analyses are flowing. But what does this mean for AGI timelines? And a lot of people in in the AI world are sort of tuning all of this out because they see recursive self improvement, AGI, ASI, the build out as more important. And I just wanted to sort of reality check the AI supply chain to understand how does oil actually affect data center construction, AI production token pricing?
Speaker 2:Like, is there any effect? My conclusion was that it's it's very moderate, but there are some interesting effects in the financial markets that are probably the bigger takeaway. It's still interesting to hear about, like, yes, oil actually does is used in the production of AI, at least a little bit. So power has been at the forefront of the AI pushback. Like, everyone's worried about these local energy prices, these electricity prices increasing near the data centers, but and it's been it's become like a political issue.
Speaker 2:But pain at the pump might become a bigger story as gasoline prices spike, And that has been paint at the pump. Oh, it's been the most tangible sign of inflation. It moves so quickly, you know, one one jitter in the economy.
Speaker 1:And it's a huge component. You know, people on the coast, people in tech don't have a good sense for this. Right? If gas gas for a lot of people, gas could quadruple
Speaker 2:Yeah.
Speaker 1:And it wouldn't they wouldn't really notice it. No. No. But if you actually look into the average American, how gasoline fits into their budget Yeah. It's a meaningful component of their monthly budget.
Speaker 1:Yep. So they feel it super intensely. And It's variable Yeah. Cost.
Speaker 2:There are there are so many different ways where a lot of Americans go on driving vacations. That obviously is directly impacted by gas prices. And then also just psychologically, there's there's something about filling up at the gas tank where you see the number ticking up, and you're doing that on a every week basis or so that it's just so visceral. It's this thing that you have to stop and then go experience and watch the money flow out of your account, like, in real time. It's very it's very visual.
Speaker 2:It's very interactive.
Speaker 1:Yeah. I remember I remember I I must have been probably 18 at this point where I would just go and I would like, for a long time, I just put I'd I'd, like, you know, prepay
Speaker 2:for a certain amount. Got $20. I got 20 I'd let's see how much I And
Speaker 1:it felt like really like, I felt like the king of the castle Exactly. Just put my car down and let it run up.
Speaker 2:Yeah. Yeah. Yeah.
Speaker 1:Yeah. Far to go.
Speaker 2:No. No. No. Totally. So pain of the pump's been like a big, broad economic issue.
Speaker 2:It's so visceral. Like, the actual phrase has a lot of power in politics and in, like, attack ads. Like, your gas prices are going up under this person. That's why you got to vote for me. This is this is common.
Speaker 2:But in AI circles, the discussion's been much more focused on RSI now is the new acronym that everyone's focusing on, not AGI. AGI is here. We know artificial general intelligence. We passed the Turing test. The models can do things generally, intelligently.
Speaker 2:But can they recursively self improve? Are they RSI? Are we in RSI now? Is it a coming? Is this going to be a fast takeoff?
Speaker 2:Is this going be a slow takeoff? Well, something's taken off. Dylan Patel said being in SF is like being in Wuhan right before the pandemic. Something is happening. It's gonna hit everywhere, but so few people know it.
Speaker 2:So he's sort of echoing that something big is happening mentality, this idea of, like, recursive self improvement being AGI pilled. A of smart people agree there's definitely something happening. There's still this question of, like, what is sticky? What where will diffusion take longer? What will be sort of AGI resistant or ASI resistant for a variety of reasons, even if they're completely irrational and you could get a better thing from a computer.
Speaker 2:But people like the human version or whatever.
Speaker 1:The irony is that George Hotts hitting the timeline to raise money Yeah. Makes me more bullish on acceleration. Yeah. Right? Because if not, he's obviously not historically been a huge fan of venture venture capital.
Speaker 2:Well, he's not raising from traditional VCs, he said. He he specifically said he wants to raise from like like basically like friends and family type of round. He does want 20,000,000. We have this in the timeline somewhere. Where is this post from George
Speaker 1:I'll pull it up if you want
Speaker 2:to keep Tiny Corp. He said that he's raising $20,000,000 to buy a building. He's he's gonna buy a building. Here it is. If so he said and this is from Tiny Corp, the makers of Tiny Grad and the Tiny Box.
Speaker 2:If Tiny Corp was raising $20,000,000 at $200,000,000 valuation, who'd be interested? Business model is basically this. Buy an $11,500,000 building with five megawatts of power, link in our Discord, wait for AMD to launch the RDNA five ninety six gigabyte cards mid two thousand twenty seven, preorder 3,000 cards. Hopefully, we can negotiate for $2,500 each, build $520,000 tiny boxes with six of the cards in in each box, run all the Chinese LLMs, make $600,000 per month revenue selling tokens on OpenRouter. Market depth is there.
Speaker 2:This is 1% of OpenRouter. Improvements to TinyGrad yield revenue improvements. Due to how power is priced in Oregon, it's only like $50,000 for the electric bill before the four megawatts, before they price for peak, not usage. We get like 3C kilowatt hour power, dollars $0.03 per kilowatt hour. We can also make $100,000 per month leasing colocation space to Comma.
Speaker 2:Building and cards paid off in three years max, investment made back, low risk of being undercut since we're using consumer GPUs and running the cheapest colo you can believe. If someone chill wants in, I'd do it. I'm not going to hype fake tech, but demand for tokens is going to skyrocket. Look at the OpenClaw install numbers. With crazy good optimizations, we could potentially get 3x more from the machines, and we have electricity for 3x more machines, 5.4 revenue per month, then continue to scale from there, custom chips, etcetera.
Speaker 2:He's starting a Neo Cloud. Are you starting a yeah. He's gonna be serving tokens. So that was that was, you know, incredibly bullish. You know, demand is definitely there.
Speaker 2:At the same time, he he sort of took the other side of, this is different. This is crazy. He says there's no AGI. There's no magic threshold. You guys see auto research change the random seed from 42 to one thirty seven and OMG, it's AGI.
Speaker 2:It's over. Yet you critique the junior engineer for the same stuff. The cost of development is falling. Overpaid engineering struggles to compete. That's the story.
Speaker 2:And he says, humanity has been recursively self improving for centuries. So everything new is old.
Speaker 1:I know. I obviously have a huge amount of respect for George.
Speaker 2:I know.
Speaker 1:But at the same time, saying there is no artificial general intelligence when you're also arguing that engineers are struggling to compete Yeah. With new technology that is replacing them. This how is this not Yeah. Artificial intelligence competing with
Speaker 6:Yeah.
Speaker 2:So so I think he is he is agreeing with that, but it's that it's that there is an immense amount of of desire for this binary moment. This is the singularity. This is API. This is ASI. The RSI is here.
Speaker 2:This thing is happening right now, and there's before and after, and everything has changed. And and and he just doesn't see it that way,
Speaker 6:I think.
Speaker 2:I think he sees it much more like the Internet, the mobile phone, like other technologies that have been rolled out
Speaker 1:Continue.
Speaker 2:Electricity. Yes. There is, like, a before and after, but you can only really, you know, define the period by maybe a decade, and you need a few decades to understand that moment. It's very hard to go back and, you know, there is, like, the iPhone moment, and there is, like, you know, the first launch of, I don't know, AOL. Like, I don't even know I don't even know what the iPhone moment of the Internet was just because it was sort of a slow rollout.
Speaker 2:Anyway, live GPU clusters. Where's Jordy going? Oh, you're moving the goalpost. Okay. Jordy's moving the goalpost.
Speaker 2:Where are you moving the goalpost to?
Speaker 1:Just over here.
Speaker 2:Okay. But why? Is there a reason?
Speaker 1:Well, mean, they have to be moved. George is moving the goalpost.
Speaker 2:Oh, okay. Okay. Okay. He's moving the goalposts. So I said, well, you know, it's clear that the AI industry continues to grow and continues to need more and more power and compute as we've seen from George Hots' new project.
Speaker 2:That means large data center campuses. But if they're not in random office buildings that George is picking up for on the cheap, they're probably going to be built with construction equipment. So what does this mean? They they don't just drop from the heavens. They require building, which requires oil.
Speaker 2:But how much oil? And is oil a serious and is a serious oil shock enough to impact the AI build out in a meaningful way? Spoiler alert. Basically, no. But live GPU clusters in The United States do not use much oil directly.
Speaker 2:Only 0.6% of US electricity in 2024 came from petroleum. We're much, much more dependent on natural gas. Something like 42% of US electricity is natural gas. And so America ramped up natural gas production significantly over the past two decades. A lot of that was in reaction to the the wars in The Middle East.
Speaker 2:Hey. We need to be less dependent on on foreign oil. We need to be energy independent. And so you have the fracking boom, the natural gas boom, and that's where a lot of our fossil fuels come from today, I believe. Data centers only consume a single digit percent of US electricity.
Speaker 2:So you're looking at point 6% of a few percent is, like, the actual impact. So the short term impact of high oil prices should be very limited on AI. Building out new capacity is
Speaker 1:much more That that said, it's it's worth noting. Think I think it's something like Qatar produces something like 20% of the world's liquefied natural gas. Right? So But not for us. Yeah.
Speaker 1:Not not for us. Yeah. So we're in a good spot. But this is gonna impact the data center build outs or or active data centers across the rest of the world Yeah. If this continues.
Speaker 2:Yeah. So when you're talking about building new capacity, building new data centers, oil is a little bit more involved. So diesel powers, trucks, trains, boats, barges, generators, pumps, compressors, excavators, and tons more construction equipment. Petroleum is also broadly used for plastics, polyurethane, and solvents that all work their way into the data center supply chain. The biggest problem is delaying already tight schedules because of narrowly available components going out of stock.
Speaker 2:So the price of oil goes up. There's one marginal factory that can't produce one ingredient that goes into the rack, and and that slows things down. You you have to wait a week while you find another supplier. That that stuff can add up to just a little bit of a delay. This happened during COVID, and the AI industry was already experiencing something similar with transformers.
Speaker 2:And so you don't want products getting stuck in transit or going out of stock. But the bigger problem and the one that people should be talking about, and I think you were debating with Dan Primak at Axios about this, is macroeconomic. So higher prices, higher oil prices lead to higher inflation. If the Fed has to raise rates to control inflation, capital formation for mega projects gets a lot harder. So JLL has this estimate.
Speaker 2:The next 100 gigawatts of data center capacity could require about $870,000,000,000 of new debt financing. And so using this rough number, every extra 50 basis points of borrowing cost on $870,000,000,000 is $4,350,000,000 in annual interest expense. And so that's a I don't know if that's a huge squeeze on cash flow, but it's certainly a squeeze on cash flow. If you're a hyperscaler and you're going gig along AI CapEx now and you're like, yeah. We're gonna do 200,000,000,000 this year.
Speaker 2:We're gonna do 500,000,000,000 next year or whatever across a couple companies. You start looking at interest expense bills that are already huge, but then they're gonna go up by 5,000,000,000 or up by a full more a full extra billion just with a 50% base 50 basis point increase, half a percent. So half a percent change in the Fed funds rate can echo through. Even a quarter percent could probably could probably be amplified into, you know, a a multibillion dollar, like, cost line on your on your balance sheet or on your income statement as you're servicing that debt. Overall, higher oil prices do make the AI rollout harder but not really for existing AI capacity.
Speaker 2:I wouldn't expect token pricing to change based on the price of oil, but there is this bigger question of macroeconomic resilience during a time where our largest tech companies are digging through the couch cushions to find every penny to win the AI future. Middle Eastern investors pulling back, I agree with you. I do think that this is a significant risk, because if they pull back on big big investments, makes these mega rounds more difficult, retail investors could fly to oil stocks or defense stocks. They think that this is, like, going to be a thing for months and months and months. It could all flip if there's a quick resolution though, which is what I'm hoping for.
Speaker 2:But if things continue, the industry walks an even tighter tightrope.
Speaker 1:So Yeah. I mean, the the big question right now is the hundreds of billions of dollars of of sovereign AI projects in The Middle East. Right? I think a lot of those people are gonna be like, do we wanna send, you know, billions of dollars of of GPUs?
Speaker 2:Over there? And then also the money coming here is Yeah. A thing where you might wanna spend it elsewhere. Quickly, let me tell you about Turbo Puffer, serverless vector and full text search, built from first principles in object storage, fast, 10 x cheaper, and extremely scalable. And let me also tell you about public, investing for those who take it seriously, stocks, options, bonds, crypto, treasuries, and more with great customer service, whatever
Speaker 1:Yes. Investing strategy. We have to say happy birthday to happy birthday to Trey.
Speaker 2:Happy birthday, Trey.
Speaker 1:Love you.
Speaker 2:Excited to have you. Thank you.
Speaker 1:One of our strongest soldiers.
Speaker 2:We appreciate you. It's been an honor And podcast. But been noticing that it's been hotter in Los Angeles? I have. Downtown Los Angeles is forecasted to approach a 100 degrees Fahrenheit on Thursday and Friday in March.
Speaker 1:Which is why
Speaker 2:Not good.
Speaker 1:We're gonna do We the have a new segment for you today. We're doing the weather on TBPN. We have our very own Ben.
Speaker 2:Hey, Ben.
Speaker 7:Hello guys. How are you?
Speaker 2:We're doing great. Tell us about the weather.
Speaker 1:What's happening?
Speaker 7:Well, wanna start off by saying, as you can see, the weather today for the low low of today is gonna be 75 degrees Fahrenheit up there. Mhmm. High of a 100 degrees Fahrenheit down there. Mhmm. But there's something I actually wanted to point out that I saw and I thought that was quite interesting.
Speaker 7:As you can see up here, there's a localized low pressure area up there and a localized high pressure area right down there if you can see that right there.
Speaker 2:Is that in rain?
Speaker 7:The issue with that's not normally an issue, not a cause for concern, and it's not very common for this time of the year. However, today, later in the afternoon, these two areas are gonna collide. They're gonna hit each other.
Speaker 2:No. Really?
Speaker 7:What that's gonna cause is a barometric pressure inversion. Okay. It might sound
Speaker 8:a little bit
Speaker 7:scary, but I guarantee there's no cause for concern. All that means is that hot air rushing in from the West is gonna collide with that cold air rushing in from the East, and it's gonna cause a bunch of turbulence in the sky moving all the airwaves around and oscillations in the sky. However, I wanna add one more point. A byproduct of this effect is that all that humidity that dropped after that hot air moved to the bottom Mhmm. Is gonna raise up because the water cycle, you know, evaporation and stuff.
Speaker 7:It's gonna raise up into the sky into those clouds. It's gonna cause big clouds in the sky. And eventually, all that water is gonna fall down onto the ground. We're gonna have big rain later in the afternoon.
Speaker 2:Wait. It it it's actually gonna rain in LA?
Speaker 7:Yeah. I I It might sound crazy, but I just want it for all you guys at home. I definitely try to step outside with a jacket today, maybe a hoodie just in
Speaker 6:case the
Speaker 1:rain comes. I'm
Speaker 2:I'm I'm fact checking you right now, and the weather app says it's gonna be sunny all week. Is this just complete fake news?
Speaker 1:My team Those
Speaker 7:are the numbers my team gave me.
Speaker 2:This is complete fake news.
Speaker 7:No. No.
Speaker 2:No. This is the fakest news I've ever heard.
Speaker 7:All that John.
Speaker 1:John. I'm I don't trust your app. I trust the weather.
Speaker 2:Apple says it's not gonna rain Yeah. Listen. The entire month. There's zero rain.
Speaker 7:Did you
Speaker 2:look at the transatlantic current? No. The Transatlantic current Zero inches today.
Speaker 1:Inches on Wednesday. John, you're really gonna you're really gonna trust him. I'm asked. You're really gonna trust him application. Yeah.
Speaker 1:It was probably vibe coded yesterday over Ben who's doing the weather.
Speaker 2:How did he get here? What happened here?
Speaker 7:Yeah. Guess. This is no. If you look right there, you can see a localized high pressure area. That's a localized
Speaker 2:I mean, this is an over eager weatherman who's just looking for drama in the most boring weather market in America, which is Los Angeles.
Speaker 1:I think I think this is the most important story in the world.
Speaker 2:This is ridiculous.
Speaker 7:You guys can look at the jet streams. They're coming in from the West.
Speaker 2:I don't wanna hear anymore mumbo jumbo about jet streams. Get out of You're done.
Speaker 1:Great work, Ben. Thank you, Ben.
Speaker 2:The fake news weather will return. Maybe never. Who knows? Let me tell you about Restream. One livestream, 30 plus destinations.
Speaker 2:If you want a multistream, go to restream.com. And let me also tell you about Gemini 3.1 Pro. With a more capable baseline, it's great for super complex tasks like visualizing difficult concepts, synthesizing data into a
Speaker 1:This is the future.
Speaker 2:Or bringing creative projects to
Speaker 6:life.
Speaker 1:This is the future of the weather. Yes. You have a weatherman
Speaker 2:Yes.
Speaker 1:Who gets into a live altercation Yes. With one of the other hosts just constantly fact checking.
Speaker 2:It's on-site next time, Ben. It's on-site. If you pull up with some fake news on TBPN's weather report, it's on-site.
Speaker 1:Ben, you crushed it.
Speaker 2:Good job. So it is it's not Adam Smith's birthday. It's the twenty fifth anniversary of the publication. Two hundred and fiftieth. Sorry.
Speaker 2:Two hundred and fiftieth anniversary of the publication of the Wealth of Nations. Tyler, you have a copy?
Speaker 9:I I have two of our copies here.
Speaker 2:You have two
Speaker 1:of our copies.
Speaker 9:So, yeah, usually I have them with me, but I didn't bring mine to work because we have a we have a few here.
Speaker 2:Yeah. Yeah. Yeah. Yeah. Obviously.
Speaker 2:For those who don't know, give us the pitch for Wealth of Nations. Why should people read Wealth of Nations in 2026?
Speaker 9:Yeah. I mean, this is kind of the OG, you know, economics book. Right?
Speaker 1:Was it it was so base like, put it into, like, AI terms. This is, like, situational awareness type of I mean, it was
Speaker 9:this is like attention is all you need.
Speaker 2:It was like the chat GBT moment for capitalism, Yes.
Speaker 9:But I I would like to, you know, I I have a few quote quotes I would like to read.
Speaker 2:Okay.
Speaker 9:Alright. Let me open the
Speaker 2:Let's see. It's a big book. It's like 800 pages.
Speaker 9:It's very long.
Speaker 2:And it's really, really dry. I mean, it reads like it was written 250 ago.
Speaker 9:Okay. So this is I I think this is kind of one one of my favorite quotes. Yeah. Kind of the, you know, seminal. Every individual necessarily labors to render the annual revenue of the society as great as he can.
Speaker 9:Mhmm. He generally, indeed, neither intends to promote public interest, nor knows how much he he is promoting it.
Speaker 2:Yes. He
Speaker 9:intends only his own gain. Yes. And he is in this, as in many other cases, led by an invisible hand to promote an end which was no part of his intention.
Speaker 2:Yeah. Rational economic actors, homo economicus, thinking of the the value for societies created by rational self interest. It's sort of the spin on, like, greed is good. It was like he was like an OG of that of that thesis. The greed maxer.
Speaker 2:Yeah. Productive labor, efficient markets, free markets, don't hold gold, don't control trade. An important read in 2026. Somewhat heretical. There are many people that disregard that book.
Speaker 2:Maybe they should not. Maybe they should give it a crack up. I think it's Evergreen. I think it's Evergreen. I mean, was that debate over, like, Tyler Cowen over the weekend, and somebody was taking shots at, stubborn attachments for being like not advancing the discipline of economics.
Speaker 2:And my pushback to that was that I actually think it is an incredibly it's just a fun read. It's like really well written and it's very entertaining and it's very thoughtful. But it's particularly interesting to hear someone sort of just rehash the tenets of liberal democratic capitalism in 2020 I think he read it in 2023, 2022, something like that. But, like, like, there there there are a lot of voices out there who are anti capitalist, anti democratic, anti liberal, and Tyler Cowen is just retreating to, like, a time tested, time honored economic philosophy that is deliberately not revolutionary. And I thought that it was a great book for that reason and that's why I recommend stubborn stubborn attachments to so many people and so widely.
Speaker 2:But if you wanna crack open the real the real the real hard stuff, pick up a copy of The Wealth of Nations. Why is one of those copies so much smaller? Is the font size just smaller?
Speaker 9:Yeah. It's super small. It's also two columns.
Speaker 2:Oh, two columns. Wow. Yeah. If you can get through that, you're you're you're truly you're truly elite in the global economy. What does the timeline have to say about the the the masterwork that is turning 02/1950?
Speaker 2:It's now needed more than ever, says the Washington Post. On 03/09/1776, four months before the American colonies broke with Britain over the issue of taxation, a little known Scottish thinker published a long dense book with an unpromising title, Shots Fire, an inquiry into the nature and causes of the wealth of nations. Two hundred and fifty years later, the Adam Smith is by any objective measure easily the most widely cited and widely quoted economist who ever lived. Astonishingly, his work still frames the central questions we face, not just about free markets, trade, and capitalism, but about the nature of human society itself and even what it is to be human at all. Smith was born in 1723 in Kirkcaldy, Scotland, educated at the universities of Glasgow in Oxford.
Speaker 2:He initially made his name not as an economist, but as a moral philosopher. His first published book, The Theory of Moral Sentiments, offered a radical theory of how we form moral judgments. Radical because it derived from the creation of moral values, not from scripture or divine grace, but from human sympathy and mutual regard. The Wealth of Nations as his second work, second major work came to be known, was an extension of that product. The book is not, as sometimes believed, a hymn to greed, a pain to market fundamentalism, or red in the tooth and claw capitalism.
Speaker 2:It was an attempt to understand how a commercial society could generate prosperity without collapsing into corruption. This twenty fifth this two hundred and fiftieth anniversary is not a moment for hagiography. It is an opportunity to recover a way of thinking that is directly relevant, indeed urgent to the economic, social, and political challenges we face today. Begin with trade. In his own time, Smith's great target was mercantilism or what he called the mercantile system, the idea that wealth consists of hoarded bullion and that trade is a zero sum contest.
Speaker 2:Governments granted monopolies, imposed tariffs, and manipulated commerce in the purported pursuit of national strength. Producers were widely protected. Consumers often ignored altogether. Against this, Smith One of
Speaker 1:these days, we should have Tyler just do a speed reading of the Wealth of Nations live on the show as long as
Speaker 2:The speed run? Yeah. Wealth of Nations speed run, any percent glitchless. You need to do glitchless, though. Glitchless.
Speaker 2:For sure, a glitchless run is the only way we can do it. No glitching. No no AI summaries. I think that would take you probably five days. Probably five days.
Speaker 2:It's so dense.
Speaker 1:Have you seen how fast you can rip a Rubik's cube?
Speaker 9:This edition is 480 pages around.
Speaker 2:Oh, I guess you don't have to read all the footnotes and stuff. You can just kinda skip through it. But it is a it is a thick book. But the but the way it's written is just really not fun to read. It's like it it reads like it's 250 years old.
Speaker 2:It's not it's not like a page turner.
Speaker 1:Well, we need to talk about the wealth.
Speaker 2:But first, vibe.co. We're DTC brands, b to b startups, and AI companies advertise on streaming TV. Pick channels, target audiences, and measure sales just like on Meta. And let's also tell you about fin.ai, the number one AI agent for customer service. If you want AI to handle your customer support, go to fin.ai.
Speaker 2:We need to
Speaker 1:talk about the wealth of Sundar Pitch I.
Speaker 2:Pitch AI. Sundar It's a new
Speaker 1:pay deal worth up to $692,000,000.
Speaker 2:Is this like 10 times what Tim Cook's making? No. It must be over over time. Right?
Speaker 1:Mean, Cook's making around 70 a year 70 a year. Combined.
Speaker 2:So if he works for ten years, he makes what Sundar makes in three? It was gonna be a
Speaker 1:Oh, is. It is a I mean, but this is what we've been we've been
Speaker 2:We've been advocating for this. So this is good.
Speaker 1:For
Speaker 2:this. Yeah. No. No. We're we're we're we're in support.
Speaker 1:Yeah. Google has increased Sundar's
Speaker 2:You know potential pay to
Speaker 1:692,000,000 over the next three years.
Speaker 2:You know that Tim Cook dropped this in the Apple board members group chat as soon as it hit. He was just like, dude, this is the this is a cool article. You should
Speaker 3:read this.
Speaker 1:Drops it in.
Speaker 2:Just drops it in. Yeah. Exactly.
Speaker 1:The bulk of his package comes in performance units Yep. With a target value of 126,000,000 split evenly into two Mhmm. Branches. The PSUs are valued by the parent company, blah blah blah. It could pay out as much as twice the target of a quarter million if it outperforms significantly or nothing if it lags behind.
Speaker 2:Gotta beat the S and P 100.
Speaker 1:Pichai will receive Waymo stock with a target stock in Waymo with a target value of a 130,000,045 million in Wing Aviation. That's their their their drone
Speaker 2:delivery platform.
Speaker 1:Platform. Again, both can pay out up to 200% of the target.
Speaker 2:Yeah. They got a lot of incentive pay. Like, if if he delivers and the and the company does well, he should he should be richly rewarded. Barron's was writing about the the death of the mag seven. Mag seven's over.
Speaker 2:Up and down Wall Street, the mag seven era ends, Barron's is calling it. They said, stick a fork in it. Turn out the lights. Hasta la vista. Say it any way you'd like.
Speaker 2:The simple truth is the mag seven trade is over. Finito. I love, Barron's writing. Dead. The collective stock market outperformance of those seven tech icons, Alphabet, amazon.com, Apple, Meta Platforms, NVIDIA, Microsoft and Tesla is now a thing of the past.
Speaker 2:The group may still do okay and some of the individual stocks may even kill it, but the slam dunk, set it and forget it, run circles around the market era of the Mag-seven is gone with the wind. Recall that BofA securities analyst, Michael Harnett, coined the term Mag-seven in 2023, pretty recently, referring to the 1970s John Sturges' Western Gunslinger flick, an adaptation of Akira Kurosawa's 1954 film Seven Samurai. Have you seen Seven Samurai? No. Have you seen The Magnificent Seven?
Speaker 1:The stocks?
Speaker 2:No. The movie. The movie. The western. No.
Speaker 2:It's Thomas Sam and Samar.
Speaker 1:In what world?
Speaker 2:Production team. Has anyone seen either of those movies?
Speaker 10:I think we watched them
Speaker 1:in school.
Speaker 2:Okay. Okay.
Speaker 1:Which ones have seen in while? You're saying didn't watch them, you studied them? Well,
Speaker 9:I haven't seen them in a
Speaker 2:while. Which ones have you seen, Tyler?
Speaker 9:Seven Samurai.
Speaker 2:Seven Samurai. It's a great one. It's a it's it's a very it's a very good movie. Anyway, Barron's is bearish on the mag seven but like coke. It does sound like coke.
Speaker 2:I think the mag seven will do.
Speaker 1:It's like the it's over for the largest and most profitable companies in history that stand to benefit Yep.
Speaker 2:Actual argument was the growth to value narrative, the loss of cash flow as they increasingly invest in CapEx. The the the financials will look very different. There'll be margin compression, that type of thing. It it is not an unreasonable take, but it is it is funny the the way it's written. Anyway, Sundar Pachai took over as CEO in August 2015.
Speaker 2:He's going on eleven years in the seat. Google's market cap has increased almost sevenfold from half 1,000,000,000,000 to 3,600,000,000,000.0, briefly topping 4,000,000,000,000 in January. This surge has made the Indian born 53 year old former McKinsey consultant a billionaire. He joined in 2004 and made his name developing the Chrome browser and leading the Android division. He had been criticized for being too slow to adopt AI at the search giant, allowing OpenAI to release the first hit product, ChatGPT, in late twenty twenty two, but has since bounced back, releasing cutting edge AI models and integrating the technology into its dominant search engine.
Speaker 2:Yes. He's done very, very well on that front. Pichai has also navigated a duo of antitrust cases brought against Google's search and app store businesses, avoiding the worst case scenario of forced breakup. A third lawsuit is pending against the advertising network. Tyler, do you think that, DeepMind counts as as RSI capable at this point?
Speaker 2:Recursive self improvement. So there was a take on the timeline where, basically, someone was making the argument that x AI is cooked because while Anthropic is using Claude to build Claude and OpenAI is using Codex to build Codex. So both of those leading private AI labs are recursively self improving. X AI was using Cloud Code to build Grok, and so they don't have the same flywheel of the tools that build the tools build the tools.
Speaker 9:Yeah. Because I think they have access to Cloud Code
Speaker 2:Yes.
Speaker 9:But then they still have access to Codecs.
Speaker 2:Okay.
Speaker 1:So
Speaker 2:Yeah. I I
Speaker 9:I assume it is helpful.
Speaker 2:It would be interesting. I feel like we don't get as many little data points from the Gemini and DeepMind team because it's a public company, and they need to, you know, maybe break out financials at some point.
Speaker 9:Yeah. I think, like, vague posting is is probably, you know, more Yeah. Heavily looked at.
Speaker 2:Exactly. Exactly. Because you would yeah. Yeah. Because it's totally possible that they're doing the same thing.
Speaker 2:I mean, Demis has been extremely just very clear about the path this goes, and he's very AGI understands the exponentials and whatnot, so he certainly messaged it at a high level without actually saying definitively yes. Like, we are at a point where we're not writing any code by hand, that type of phrasing that you see in the RSI crowd. Anyway, Pichai got last got a stock award in December 2022 worth 218,000,000, which was structured in the same way. His earnings are topped up by his personal security costs, which rose to 8,300,000.0 in 2024. Earlier in the week, he sold 32,500 class c shares at an average price of $303 worth roughly $10,000,000.
Speaker 2:The Bloomberg Billionaires Index estimates that he has sold about $650,000,000 in stock since becoming CEO. They still own along with his wife, he owns one point seven one point six seven million shares of Google worth half 1,000,000,000 at the latest stock price. And Google's founders Sergey Brin and Larry Page still control the company through their ownership of super voting class b stock, which gives them 56% of decision making power. Interesting. Think you gotta give Miami
Speaker 1:paper hands? Oh. He sold 650,000,000 stock?
Speaker 2:No. No. No. He he sold that, but he's, oh, and he only owns half 1,000,000,000. So, yeah, he sold more
Speaker 1:than Hopefully hopefully hopefully he put it all
Speaker 2:into shy. But he's getting topped up and he's diamond handing is the new he's handing.
Speaker 1:I'm just saying. I'm just saying, looking at the stock charts since 2015 when he became CEO Yeah. Would have done pretty well just
Speaker 2:to This
Speaker 1:whole not do anything and trust trust his own process.
Speaker 2:Hey. Hey. Maybe he has amazing car collection or something. You know? There's a minute.
Speaker 2:There's a of things.
Speaker 1:Metal, man.
Speaker 2:You can't drive a Google class class a share. You can't you can't sleep on a bed of class c shares, Jordy. You gotta you gotta get liquid. No. You gotta give Miami Tech its due because everyone's saying, oh, Miami Tech, they don't have any startups.
Speaker 2:There's no funding there. There's no Miami Tech isn't panning out. Well, Google is now counts as a Miami tech company because the Google founders bought houses there. You also got to count Meta as a Miami tech company because Mark Zuckerberg has a house there. So you put Meta and Google, that's like $6,000,000,000,000 in market cap for tech companies I in count it.
Speaker 2:I count it. Anyway,
Speaker 1:moving. Owner of Four Loco is exploring a sale of a of a storied alcohol brand sources say.
Speaker 2:Wow. Let's dig into this, but first let me tell you about Plaid. Plaid powers the apps you use to spend, save, borrow, and invest securely connecting bank accounts to move money, fight fraud, and improve lending now with AI. And let me also tell you about Cognition. They're the makers of Devon, the AI software engineer.
Speaker 2:Crush your backlog with your personal AI engineering team.
Speaker 1:So The parent company of Four Loco, the canned alcoholic beverage that became a college campus sensation in the late two thousands Mhmm. Before being reformulated under regulatory pressure is exploring. I cannot believe they nerfed Four Loco. It's It was too powerful. Was too Like, you you can just imagine the trajectory of The United States if it hadn't been nerfed and then
Speaker 2:Straight downward.
Speaker 1:Really?
Speaker 2:Yes. It was so bad. It was so impact I thought people
Speaker 9:were having like heart attacks.
Speaker 2:Yes. It was terrible.
Speaker 1:So Not really?
Speaker 2:Four Loco was a 24 ounce can, so twice the size of a normal alcoholic beverage.
Speaker 1:Trey, the birthday boy says it was amazing.
Speaker 2:It was amazing. Okay. Now we have buzz balls. That's true. So, Four Loco, the original formulation was twice as big as a normal can of Bud Light or Miller Lite or Coors Light, your normal beer, something that you would grab.
Speaker 2:So it two drinks that way. And then instead of it being 3.2% alcohol or 4.5% alcohol, it was like 10% alcohol. It was the strength of like wine almost. And so a single Four Loco was like four or five beers in one can. And then they also added like two hundred milligrams of caffeine.
Speaker 2:So you were you would you would become incredibly intoxicated and inebriated, but then also incredibly stimulated from all the caffeine. And that spelled doom for many people. People would be very high functioning but completely inebriated and and discombobulated. And so they would, know, get into all sorts of trouble whether or not it was behind a motor vehicle or
Speaker 1:discombobulator.
Speaker 2:It was the discombobulator. It was also wildly illegal from an FDA perspective. You can't mix alcohol and caffeine in a single product. That's just a rule that they chose not to follow and they figured out how to make it. And then the FDA eventually gave them a warning letter and they pulled the caffeine out and that, of course, killed the viral sensation that was four Loco.
Speaker 2:But apparently, the company continued to sell products that were compliant with the current FDA regulation. Nerf. And that's nerfed. The nerf four loco continued to
Speaker 1:Somehow, they're looking to and potentially will capture somewhere around 400,000,000 on the sale.
Speaker 2:The brand value. So people it's hilarious that they're working with JPMorgan on this. That's wild. So Fusion Products is the owner of the brand. They're working with JPMorgan on the sale process The potential sale underscores how ready to drink beverages have emerged as a growth category in an otherwise sluggish alcohol market.
Speaker 2:US beer and wine US beer, wine, and spirit sales declined in 2025. We've talked to a number of guests about the declining alcohol sales as folks move over to more functional beverages.
Speaker 1:The CMO had to say. For over a decade, our sales have been the leading flavored malted have have been leading the flavored malted beverage market by embracing bold innovation, unconventional marketing, and a risk taking attitude that delivers results year after year.
Speaker 2:I guess they kept building. They also own
Speaker 1:The parent company also owns Mamitas Pirate Water Pirate Water. Basico Tequila. What is Earth and just Earthquake.
Speaker 2:Earthquake. Earthquake seems like something you definitely want
Speaker 1:From the makers of Four Loco, Earthquake.
Speaker 2:Pirate water is a wild name for a brand too. I mean, it's a distinctive can. It looks remarkable. It jumps off the shelves. When these when the Four Loco nerf went into effect and they pulled them from the shelves because before they reformulated without the caffeine, they had to pull them from the shelves entirely.
Speaker 2:They were a hot commodity. And folks would go around stockpiling them before they were banned and and they were the one of the most guarded resources in all of college campuses. Wow.
Speaker 1:Earthquake is just a can that's just 10% alcohol.
Speaker 2:Wow. Do not recommend. Very
Speaker 1:risky. Never quake.
Speaker 2:Anyway, in 2010, the FDA sent warning letters to caffeinated alcohol beverage makers, including Four Loco, saying that the caffeine in their products constituted an unsafe food additive. So the FDA has specific categories for regulation around around what is an alcoholic beverage, what is a caffeinated beverage, and they'd never the two shall mix. So they don't let you put these two things together. Of course, vodka Red Bull was the precursor to this where folks would order a Red Bull and then they would put vodka in it, I think, but at a bar, which is made at the time of sale. And so that was legal, I believe.
Speaker 2:And this might be fake news, but I remember hearing a story that at one point, Red Bull was sued because a gentleman had purchased a series of vodka Red Bulls, become incredibly inebriated, gotten behind the wheel of the car of a car, got into some sort of accident and wound up in front of a judge. And the argument was that he wanted Red Bull to be held liable for his actions while inebriated behind the motor vehicle. And his argument was that it was only because of the product Red Bull had been mixed with the vodka that he made the decision that he was alert enough, stimulated enough from the caffeine to get behind the car wheel in the first place. And this might be apocryphal, but the story goes that Red Bull made the argument that the caffeine actually made him more capable to drive, not less capable to drive, and actually increased the safety of the of the driver in the vehicle, and so they were held not responsible. But I don't know.
Speaker 2:It's just like an old wives tale in CPG. I don't know if it's true. Anyway, let's move on to phantom cash. Fund your wallet without exchanges or middlemen and spend with the phantom card. And let me also tell you about Labelbox.
Speaker 2:RL environments, voice, robotics, evals, and expert human data, Labelbox is the data factory behind the world's leading AI teams. So we're all preparing for the singularity. What is this, doing dishes? Just like, oh, if you're if you're doing dishes, you're not you're you're not worried about Dylan Patel's, you know, viral post saying that being in SF is like being in Wuhan right before the pandemic. Something is happening.
Speaker 2:It's gonna hit everywhere, but so few people know it. And that is the question of of are we in a software only singularity where the the dishwasher is life is normal? And and and every time I go to the COVID analogy, I start a clock because I remember tracking COVID in January, and I remember the viral post biology posted in February going viral. And then in March, the NBA shut down, and and in April, everyone was locked down. And so every time I see one of these, it's like COVID.
Speaker 2:I'm like, okay. Let's see what it let let's see let's see what it's like in eight weeks. Like, eight weeks is not that long. In eight weeks, if if I if I go outside and things look radically different and you talk to random people and it's wildly different, then we're like, okay. Yes.
Speaker 2:That analogy was correct. But if it's more like in eight weeks, we're like, oh, yeah. Like, you know, technology is getting more exciting and, like, tech companies are changing and, like, the market's moving. But, like, you know, the the the the the the dishwasher is unaffected. The the the pizza place on the corner is unaffected.
Speaker 2:That that I know. That's always how I I I I just think people underrate, like, how crazy a global plague was. Like, it's literally, like, once in a 100, like, the most insane thing ever. It's a the the craziest change to literally everything from the sports that you watch. You couldn't go to the movie theater.
Speaker 2:You couldn't go to dinner. Like, was changed for like a full year. It was like complete upheaval in our society and might happen. But eight weeks? Eight weeks is a little fast for me.
Speaker 1:And if you live in LA, probably still be able to go to the beach.
Speaker 2:I think so.
Speaker 1:So that's
Speaker 2:But maybe not. Maybe everything's did
Speaker 1:have a funny moment yesterday. Was telling you, but I'll I'll share it again. I was talking with my my car detailer who's become a dear friend and he was telling me that the people in my area Mhmm. Cancel their appointments Mhmm. At a much higher rate
Speaker 2:Mhmm.
Speaker 1:Than the people in the rest of LA. And he was talking about running analytics. IMessage and I was like, oh, how are you how do you do how how do you do that? And he's like, he's fully running OpenClaw. So he's Okay.
Speaker 1:Now running using OpenClaw to his, like, card detailing business. Like, yeah, just just always running in the background. Yep. It's it's some to him, he was like, it's some combination of having like an EA Yep. Or a CEO.
Speaker 1:It's like helping me manage the calendar, helping with comms, helping me make smarter decisions on routing and scheduling and all these things.
Speaker 2:So when you book a detailing, meaning that this gentleman is coming to your house and cleaning your car, you send an iMessage to him.
Speaker 1:Correct?
Speaker 2:And that is what is so interesting about OpenClaw, is that there were probably a thousand companies that were like, we're vertical SaaS for auto detailing. You got to do everything over email. And they just got eaten alive because people use iMessage, and iMessage has been this closed ecosystem that private companies could never access before. But for but it makes so much sense for him to say, hey. Run through all my iMessages.
Speaker 2:And if I'm talking to my friends about the the the the the f one race, just discard that. But if I'm talking to a client about potentially booking a detailing, put that in a spreadsheet, organize that, and then run some analytics and tell me what what the patterns are. And that's just something that should have been SaaS for a decade, but Apple has a walled garden that has now been smashed down by OpenClaw. I think it's really It and I think that I think that that's what we're seeing the first, like, use cases for OpenClaw. It's like things that could not happen for business reasons, for logical reasons, sometimes for legal reasons, if you're talking about downloading illegal films like the Napster analogy.
Speaker 2:But a lot of times, it's just it's just Apple didn't wanna do a deal to integrate iMessage into Salesforce.
Speaker 1:The way to break down the walled garden was just to have a digital guy.
Speaker 2:Basically. Yeah. Who just who just combs over all this. So, is that a new type of software that that should have existed that now is unlocked? Is it disruptive?
Speaker 2:Like, it's not like he it's not like he stopped using his previous CRM, probably didn't have anything. So now he has something?
Speaker 1:Yeah. He he's pretty tech forward. He's tried every single SaaS tool Terrible. That there is. But, like, at the end of the day, having, like, a digital guy that has access to all of your tools
Speaker 2:Much better.
Speaker 1:Is just a much better Yeah. Experience.
Speaker 2:And so I wanna know what his bill is.
Speaker 1:This month He he said he spent a decent amount
Speaker 3:Okay. On
Speaker 2:I wanna know what it is this month, and then I wanna know what it is in, a year. Because sometimes the maintenance of these systems can be really, really big. And then there's a question about, are we gonna see per token deflation, or are we gonna see prices are prices being artificially held low because of competitive dynamics? So in a year, will he be spending more or less? Will he still be getting the value?
Speaker 2:These are all interesting questions, but I'm just excited that a a digital product now exists that could not or didn't exist before, and that feels incredibly positive and cool. So very exciting. Anyway, we have our first guest of the show in the Restream waiting room. Let me tell you about console.com. Console builds AI agents that are, I mean, 70% of IT, HR, and finance support, giving employees instant resolution for access requests and password resets.
Speaker 2:And without further ado, doctor Alex Wissner-Gross, welcome to the show. How are you doing?
Speaker 1:What's happening?
Speaker 5:Doing really well. Thank you, John. Thank you, Jordy.
Speaker 2:Thank you so much for taking the time to come chat with us. Congratulations on all the virality. For those who haven't followed along so far, can you introduce yourself and the project broadly? And then I have a ton of questions, but I'll let you kick it off with an intro.
Speaker 5:Absolutely. So quick background on on myself. Originally from New York, undergrad MIT, studied physics, electrical engineering, computer science, math, PhD, physics, Harvard. Was there anything you
Speaker 2:didn't study? You studied everything.
Speaker 5:Didn't study the humanities.
Speaker 2:Okay.
Speaker 5:I've considered considered going to Cambridge to study chemistry and biology for a while. Just started it. Okay. Was told physics is a game for the young.
Speaker 2:Okay.
Speaker 5:So stayed in The US. Did my PhD in physics, focusing on AI and nanotech. Cool. I've started, invested in, advised, managed dozens of companies at this point. I've had a number of exits.
Speaker 5:And at at this point, most of my time is focused on smoothing out the singularity as as you guys were just discussing. So most recently, as as you were gesturing, a company that I helped found, Eoghan Systems Public Benefit Corporation, just announced what we've been characterizing as the first multi behavior upload, if you will, more on that in a second, of a fruit fly, which we think is a major step forward for the field. And and quite frankly, in an era when I I do another podcast, the Moonshots podcast Yeah. With my friend Peter. I I talk all the time about tiling the earth with compute.
Speaker 5:Right now, all of this AI infra build out that we're doing to the tunes of trillions of dollars of CapEx, all of this is going to artificial minds. And Eoghan is playing, I think, a very important role in doing some early pioneering experiments and developments to try to level the playing field of the future, not just artificial minds playing on all of these trillions of dollars of infra and data centers, but enabling emulations and, if you will, uploads of human minds, nonhuman animal minds at the moment, starting with fruit flies and aspirationally working towards mice and humans and getting getting the brains of a variety of human and nonhuman animals to operate in the cloud.
Speaker 2:Mhmm. So walk us through the fruit fly experiment, why it was so exciting for you.
Speaker 5:It's really interesting. If you think back remember back with the launch of ChatGPT, that was just a side experiment. That that wasn't the main focus of OpenAI at the time. Some have characterized that as an unhobbling, if you will. Yeah.
Speaker 5:GPT three, the the leadership of OpenAI thought at the time was sort of the the main entree, if if you will. Similarly, within Eoghan, I I don't mind saying this was more of a side project that the we thought it would be somewhat interesting to to the world, but I I think the the level of response and virality probably even took most of Eoghan by surprise. So this this
Speaker 2:is a
Speaker 5:video demonstration of an internal project to take a bunch of building blocks that were already sitting around for the past one to two years and put them together for the first time. So one of these was a paper that senior scientist at Eoghan, Philip Chew, published in Nature in 2024, demonstrating that just from an emulation perspective that you could take the Flywire, which was a well funded project to capture the connectome of the fruit fly brain, that you could emulate certain circuits within the fruit fly brain. We've had that for a year and a half or so. Another project, NeuroMechFly version two, it's an excellent project to be able to formulate simulations in realistic mechanical environments of fruit flies. Another project from Erzdal et al to formulate coordinated motor actuation of fruit flies, and then AI.
Speaker 5:So in summary, Eoghan was really the first to put all these building blocks together that were sort of lying around. This is actually tiny subset of what Eoghan's working on. Eoghan is working on scanning the connectomes of of human and nonhuman animals at scale and achieving whole brain emulation, which is the the holy grail of the field. There have been many projects in the past that have attempted human whole brain emulation. Eoghan is leveraging advances in AI, expansion microscopy, a variety of other techniques, but it was really just putting the pieces together.
Speaker 5:So we're quite quite honestly astonished by the the level of interest in just this one demo.
Speaker 2:I love it. I have a million questions. Let's start with IO, like the inputs and outputs. Like, what are you simulating? When I pull up the sims, that is somewhat of a simulation of a human.
Speaker 2:The you know, a a basic four loop over a bunch of random probabilities to say, are you hungry? If hungry, eat. Go over to the stove. Cook food. That's obviously nowhere near the level of detail at which a human operates.
Speaker 2:Is there an abstraction layer here where the output is quantized to things that you would think a fly would do, like move a move a wing? Or or are you operating at the level of, like, choose to fly? Or how at what level of abstraction is the fly's output happening?
Speaker 5:Yeah. This is what I was gesturing at a moment ago. So it was a paper by Erzdahl et al that bundles motor actuations
Speaker 2:Mhmm.
Speaker 5:In terms of high level representations.
Speaker 2:Got it.
Speaker 5:So a lot of people sort of the the people who got perhaps most excited Disruption in the neuroscience community about this video, some of them I
Speaker 1:think didn't bother to read Urs
Speaker 5:Sundar's excellent paper. So a lot of those this is why I'm saying a lot of the pieces were sort of lying on the ground waiting for us to pick up and and put together. So these these, like, ab these abstract motor movements were already available. Just no one had ever, prior to this, bothered to wire them up.
Speaker 2:That's cool. Then what is the do you need a system prompt? Like, do you need, like, an in like, an initialization function? Or once you once you synthesize the connectome, synthesize the connections in the brain, things just start firing. Do you have to trigger something to to start the process of consciousness or whatever you call whatever the fly is doing?
Speaker 5:No system prompt needed. On the other hand, in in the early days of ChatGPT before instruction tuning, there was no system prompt either. Yeah. In the days of of early language models, no one had thought of prompt engineering.
Speaker 10:So it
Speaker 5:it it's entirely possible that that maybe in the future, some sort of prompt engineering equivalent for for whole brain emulation will be essential, but at the moment, there there's no direct analog of that.
Speaker 2:Help me compare some of the scales that we're talking about on complexity of a fly, complexity of a mouse, complexity of a human. I imagine that we're on some sort of exponential here, but how many orders of magnitude in complexity, size, number of neurons, something like that, whatever the metric is, how far away are we from humanity?
Speaker 5:We're many orders of magnitude away. So the canonical estimate for the human brain is, depending on how you count brain cells, there are multiple types of brain cells, call it order of magnitude 100 to 200,000,000,000 cells in the brain. We're many, many orders of magnitude many orders of many orders of magnitude away from that. So this isn't going to happen immediately. We're not going to get human whole brain emulation
Speaker 2:Yeah.
Speaker 5:Immediately. On the other hand, I I think it's important I think back to the early days of driverless cars and Yeah. Autonomous vehicles. And what the world is missing right now is, I think, sort of a scale or a framework for starting to think about what that future looks like. We're at the earliest stages of that, call it level one or level two uploading.
Speaker 5:A lot of people even get triggered by the term uploading. They prefer something else. And I I think we need to start having these conversations in a critical, thoughtful way and start to define what is a framework for multiple levels of fidelity for uploading or emulation or maybe we come up with a new term. But right now, we're starting that discussion, and we're trying to do it in a thoughtful, responsible way.
Speaker 2:So you like the term uploading because you did not you did not choose a random structure or an average structure of this particular fly's brain. You chose the exact structure of a particular fly's brain?
Speaker 5:This is a very expensive structure of a very particular fry's fly's brain. This was the the structure was again captured through the the Flywire project at great expense.
Speaker 2:Mhmm. So that's like a clone digital twin type of thing. Right? Is
Speaker 5:that Aspirationally. I mean, again, we're at the earliest stages. I I don't want to oversell. Yeah. This is just a prototype, and and quite frankly, this is a sliver of everything that Eoghan is doing, but the the fact that Eoghan was able to put a few of these pieces together Mhmm.
Speaker 5:Build a prototype with relatively small effort suggests that, you know, we we talk in the AI community all the time about overhangs. Mhmm. Many people thought that the arrival of large language models was in some sense a technological overhang, that we could have one could imagine a thought experiment. Could we have had large language models twenty years ago if we simply knew what we should have been building? I I think I probably
Speaker 2:thought we were very GPU gated on that.
Speaker 5:Well, the GPUs were being spent on video games for for the first few decades. So we we could have, in principle, like, Markov models. We we knew how to build Markov Babblers. I was building Markov Babblers in the nineties and early two thousands.
Speaker 2:Yeah. We could have built if you even if you try and port back Frontier models now to a, you know, ten eighty TI, which was a graphics card for gaming, you're gonna have a bad time.
Speaker 5:Well, I mean, we're with the rise of SLMs and some of the the amazing deflationary properties of algorithmic progress in in SLMs Yeah. I think we will find ourselves in the world next two years where some of the earliest PCs could probably have hosted nontrivial conversations with LLMs. There's just that much algorithmic progress.
Speaker 2:Extremely bullish for Apple. Anyway, not to put it in hardware terms.
Speaker 1:What do you wanna do what do you wanna do next? Like, where do you where do you go from here? I know it's a I know it's a scientific More
Speaker 2:flies or straight to mouse?
Speaker 5:It's such a profoundly interesting question. So Eoghan is beginning fundraising. Eoghan wants to to tackle both mice and men. And I I think the time scale will be determined by both technical capability increases and also fundraising abilities. So I I think it's an interesting time.
Speaker 5:I I I like to say, you know, on my other pod that the singularity looks like all sci fi tropes happening everywhere all at once. And one of these one of these sci fi tropes is most definitely mind uploading, and I I'm doing my best vis a vis Eoghan systems to push that part a little bit to the left so that the AIs don't have the Dyson Swarm all to themselves.
Speaker 2:Interesting. Architecturally, what does the neural network look like that you've built for the fly? Is it is it similar to a large language model like transformer based architecture? Architecture? Or are you a beneficiary of
Speaker 4:the collection?
Speaker 2:Are is all you need?
Speaker 5:To to totally unlike the transformers. So transformer. So transformer architecture, if you subtract off the encoder and decoder layers, looks it can comes in many variants, but the most vanilla variant looks like alternating linear and attention layers.
Speaker 3:Mhmm.
Speaker 5:So totally unlike the transformer architecture, at least if if you look at it, it looks like more like a a little bit more like a graph neural network, but really that that's such an AI way of framing it. It looks like a leaky integrate and fire, a LIF model of that that we've had for for decades from the neuroscience world. So it's just a graph of nodes that have leaky entry and fire dynamics, they're firing at each other.
Speaker 2:So what does that tell you about the nature of synthetic human intelligence? Like, do you think that there is a path to human level AGI that does not involve the transformer and instead falls along your path and and your architecture?
Speaker 5:Oh, sure. Well, first of all, I I would argue we've had AGI for at least six years now, five to six years at at the very latest. Since since summer since 2020 when OpenAI published their language models or few shot learners paper, yeah, GPT three. Yep. I would argue that was an AGI.
Speaker 6:Okay.
Speaker 5:So I I I think there are many ways my bet is there are many ways to achieve generality of intelligence. I published a paper a number of years ago arguing that not only is intelligent behavior a general process, I I went further and argued that it is a general physical process that you can even formulate intelligent behavior in pure thermodynamic terms. So my bet is intelligence is this very, very general effect and lots of ways to implement it.
Speaker 2:Interesting. Do you so do you I mean, you've mentioned AGI six years ago. Singularity here, and you're smoothing it. Do you have other sort of binary benchmarks that you're looking towards, like ASI, recursive self improvement? Are any of these terms useful to you?
Speaker 5:Oh, we're already there. I mean, we're we're already error in the era of recursive self improvement. All of the Frontier Labs are are pretty public about it. Mhmm. At this point, I'm looking past the singularity.
Speaker 5:So I spend most of my time on bets for what the post singularity world looks like. So Eoghan is one of those bets. Another bet, I have a company, Physical Superintelligence, that's trying to solve all of physics with AI and doing an amazing job. I just wrote a book with Peter called Solve Everything arguing that entire disciplines are going to get steamrolled by superintelligence and that what matters now is what disciplines we aim the superintelligence at.
Speaker 1:Interesting. What what decisions do you make in your personal life in the way that you live Yeah. That are based around your beliefs around technological progress?
Speaker 5:There are a few that come to mind. So one is I'm hoping not to die. It would be a shame if if I died sometime soon, get to miss the most exciting developments.
Speaker 2:Same.
Speaker 5:That's what gets the applause?
Speaker 2:That's great. Yeah. That's that's a
Speaker 5:white pill. White pill. Very good. Brian Johnson thanks me in advance. So but I I think also, I mean, there there are so many other angles.
Speaker 5:And another is not trying to bet against the collective intelligence of the market. I've had a a number of friends who who thought they could outsmart all of the AI algo traders and day trade, and they they when they inevitably fail, they attribute it to bad luck or or something other than than just fundamentally betting against progress and betting against AI intelligence. So trying not to make the mistake of betting against collective intelligence of civilization and AI capabilities. I I also I'm I'm trying to make bets assuming that the AI capabilities keep increasing to the point where they're ontologically shocking and trying not to duplicate effort of Frontier Labs. And I I think that's perhaps something I don't see enough of in the venture community.
Speaker 5:So many new startups being formed now just aren't being ambitious enough. I have startups that newly formed or recently formed that are literally trying to grow new islands and new coastlines with AI. That would have been unthinkable years ago. I I
Speaker 1:have What is that? That's that's with physical AI, or or they're just dreaming him up?
Speaker 5:It's with AI reaching into the physical world to to steer ocean currents and grow new islands and new coastlines. That would not have been possible a few years ago. That's now starting to become possible. I have a startup that's working on solving interspecies communication starting with dogs.
Speaker 2:Love
Speaker 5:that. That would not have been possible a few years ago. So I I spend a lot of my time thinking about what the post singularity state of the world looks like and how to smooth that out and bring it here sooner.
Speaker 2:Yeah. What does the post singularity world look like where you are able to upload yourself? I assume that feels like there's a copy of Alex in the cloud, and then you still exist and not in the cloud. How do you think that that interfaces?
Speaker 5:In in my mind, if if we wind up having the copy question Yeah. Come about, that's almost a failure mode. Okay. I speaking for myself, I don't want a copy of myself. Yeah.
Speaker 3:Of course.
Speaker 5:I I want myself.
Speaker 2:Yeah.
Speaker 5:And Hans Moravec and others have written about this. What I would like to see in this space and what Eoghan is working toward, at least aspirationally, is this idea of a continuous transfer of consciousness so that it really is you. It's not a copy of you. It's not a low fidelity facsimile of you. It should be a better expanded version of you that's still you.
Speaker 5:And whether that looks like moving from what Eoghan's doing right now Yeah. With fruit flies to maybe replacing a single cell in a brain at a time, invasively or non invasively, with a substrate independent or substrate migrated implementation. Some variant of that, I suspect, is where all of this will go, so that you never have to worry, is it really myself or not? It will, by construction, be yourself. And I'm I'm not sure of the precise timetable
Speaker 2:Yeah.
Speaker 5:But I I think five, ten years from now, I think the world will see marked progress on the the problem of not just whole brain emulation, but also transfer of human intelligence to new substrates.
Speaker 2:That's very exciting. This is
Speaker 1:The new sci fi corner. I love it.
Speaker 5:We'll And write a sci fi trope everywhere all at once.
Speaker 1:Time travel? That's a good framing.
Speaker 2:Is time travel on the table?
Speaker 5:We don't know yet. Okay. So it's an interesting question. We don't know whether the physics of our universe are compatible with, I think, what most folks would construe as as time travel. There are there are versions of temporal non locality that are consistent with the physics that we have right now, but not in a useful way that you would that would be worth filming a sci fi movie over.
Speaker 2:What about faster than light travel?
Speaker 5:We don't know yet. Yeah. That is the same problem. In in some sense, exactly the same problem as time travel. Yeah.
Speaker 5:We don't know whether the physics of our universe will ultimately allow that.
Speaker 2:Because it seems like if we don't get FTL, like, there's a it's gonna be a boring travel to go to another solar system.
Speaker 5:I I will say this. I think odds are pretty good that AI in the next few years will tell us whether the the laws of physics of our universe are compatible with that or not and help us solve this. One of the reasons why I helped found physical superintelligence to to discover any of such laws.
Speaker 2:Very cool. Well, thank you so much for taking the time to come chat with us.
Speaker 1:Yeah. Great to meet you.
Speaker 2:Congratulations on the progress.
Speaker 1:Fascinating stuff.
Speaker 5:Thank you. Likewise.
Speaker 2:We'll talk to you soon. Cheers. Have a good one. Let me tell you about Figma. No matter where your idea starts, Figma make, blog code, codex or a sketch, the Figma canvas is where ideas connect and products take shape, build in the right direction with Figma.
Speaker 2:And let me also tell you about Cisco. Critical infrastructure for the AI era unlock seamless real time experiences and new value with Cisco. And without further ado, have Charles Lamanna from Microsoft in the newsroom waiting room. Now he's in the TV room. Charles, how are you doing?
Speaker 8:Good. Thanks for having me on, guys.
Speaker 2:Thanks so much for coming on the show. Why don't you give us a little bit of an introduction on yourself and then we can get into the news.
Speaker 8:Sure thing. I'm on the product engineering side at Microsoft. I my team builds out things like agents, a bunch of the the platform work and our business applications and and some low code stuff. So a bunch of odds and ends.
Speaker 2:And what what's the latest and greatest in your world?
Speaker 8:Yes. A lot of big news today. So today we announced Copilot co work.
Speaker 3:Mhmm.
Speaker 8:Which really takes Copilot beyond chat, so it can take actions and automate tasks and take delegation, basically. We had some new agentic capabilities in the Office apps and a bunch of agent management solutions.
Speaker 2:Yeah. How are you thinking about the the OpenClaw boom people writing code locally versus Microsoft has Azure? Like, it's so easy to throw everything in the cloud. And I feel like one of the barriers that a lot of people have is, like, do I need to go get a Mac mini? Can I just delegate to the cloud?
Speaker 2:Some of the apps don't just, by default, build an app in the cloud or they don't have functionality. How are you feeling about that trade off these days?
Speaker 8:Yeah. I mean, we're super excited by OpenClaw and all that means.
Speaker 2:I think
Speaker 8:it's like the resurgence of the PC. Like, what's the best thing you can give an agent? A computer. Yeah. You know?
Speaker 8:Let it run the apps. Let it log in as you. Let it message as you. I think, like, a Windows PC in the cloud is probably the main way people will ultimately use something like OpenClaw at work.
Speaker 2:Yeah. How do you think about model selection? Obviously, Microsoft has a has a time honored partnership with OpenAI at this point, but, you know, Microsoft has their own AI lab, and you can get pretty much any model on Azure. Do customers wanna pick models? How do you think about what is what's the best tool for the job in the products that you're building?
Speaker 2:But then also, how much do you wanna give to the user? Because there's sometimes when I go into a chat app and I'm like, I just want auto. You should be figuring this out. I I know the details of pro versus thinking, but I don't wanna think at all.
Speaker 8:Yeah. So that's one of the things that we think is super important for Copilot and a lot of our products the fact that it's multi model. And the reason is because the best model for a given kind of task is not always the same, and it's not even the same from month to month or quarter to quarter. Like, if I look at Microsoft three sixty five Copilot, chat is like the main way people interact with it. Our auto router is gonna put you in OpenAI models because that's what we think works best with our work grounding and our work context.
Speaker 8:Sure. But then if you go to co work, the new announcement today, and you fire off a task you want it to run for a long period of time in the background, use a bunch of tools, maybe generate some code, use like the terminal and stuff, that uses the clot family of models. And I go to Copilot one place and get both, and then we're gonna keep adding more models, models like from MAI and other places where it makes sense. But Yeah. It's really about the experience the user wants, not so much picking exactly the version or the type.
Speaker 2:Yeah. How have conversations been with business leaders in terms of deploying a product like Copilot Cowork because the the the the the ramp up for software engineers was smooth, and everyone was aware of it, tracking everything, and going from, you know, better autocomplete to agentic coding was was pretty intelligible at every step of the way for most software developers. But for someone who's lived their life in spreadsheets and email and maybe not really thought about building automations all that often, what does actual deployment of these tools look like?
Speaker 8:Yeah. It's such a great question, because I think there's like two pieces to it. The first is you have IT teams Mhmm. Which are nervous with if you imagine every single employee spinning up a whole bunch of agents that can create documents, send emails, run workflows. So they wanna make sure they have the right governance and controls in place.
Speaker 2:Mhmm.
Speaker 8:That's why another big part of the announcement today was something called agent three sixty five, which makes it easy for IT teams to manage and deploy agents from Microsoft and other companies. So that's one part. And then you have the end users. How do you teach a typical Excel or PowerPoint or Outlook user this idea of agentic working? Even in coders, it it's taken time, and it's probably not fully diffused.
Speaker 8:So that's the thing that we think we're in a really unique position at Microsoft in terms of being able to do because we can bring things like co work to users where they already spend their time inside of the Office apps, inside of Windows, and increasingly inside the Copilot app. Mhmm. And we can present it in a way where it's digestible to hundreds of millions of Office users, not just tens of millions of coders.
Speaker 2:Is the majority of your business model still seat based at this point? Because I imagine that there will be a transition at some point. How much of that transition is already happening?
Speaker 8:Yeah. We're we're still very seat based on Yeah. Like the m three sixty five side, but like, the good news at Microsoft, we kinda know how to do consumption businesses Sure. Because we got Azure. Yeah.
Speaker 8:We got a bunch of other very large capacity businesses. So, the fact that we have user licensing and then consumption and pay as you go licensing for capacity Mhmm. All integrated with one commercial kind of vehicle with customers, we think we're in a pretty good position to make that transition over the next couple of years.
Speaker 2:There was some, there was some chatter on X this weekend over, how different labs are pricing tokens and plans potentially to take market share from each other. There's a little bit of a capital fight playing out. How are you communicating Yeah.
Speaker 1:More specifically, with the application layer.
Speaker 2:Yes. Yeah. And so and so my question is, how are you talking to business leaders, CTOs about how they should think about forecasting token budgets and spend and compute budgets. If they if they're used to, okay. I have licenses and licenses maybe go up in price and my head count goes up, and I'm used to that.
Speaker 2:But all of a sudden, it's like, well, everyone's paying $200 a month, but it might be $5,000 a month in a couple months. But also, the models are getting cheaper, and the inference is getting cheaper, and the capabilities are going up, and it's very hard to predict what are you talking to what are you telling to business leaders and, CTOs? Yeah.
Speaker 8:Yeah. So I the the first thing is you have to embrace this technology, otherwise you're gonna be left behind right now. Because the productivity gains are just so significant. And if you go talk to a coder Yeah. And you told them, hey, I'm gonna take away like GitHub Copilot and the agent coding capabilities, they'd be like, I refuse to work in this environment.
Speaker 8:Like, it's just it's just inhumane almost. So Yeah. So I think the the same type of thing is gonna happen for all information work, all office work. So, nine months from now, I think if you went to somebody and said, we're gonna take away your agentic tools like Copilot, Cowork, they'd be like, no way. I'm not gonna go back to the old way of working.
Speaker 8:So so I think the first is that there's a degree of inevitability because the benefit is so large and there's such strong pull from the end users. Mhmm. That said, you everybody has budgets. Yeah. So what you're gonna see and what you're already starting to see is a bunch of controls where IT teams can start to manage exactly how many tokens or how much dollars each user can spend, and they're gonna allocate that in a way based on where they think they're getting return and diminishing return.
Speaker 8:And it'll probably be like a expense report approval hierarchy type thing. So you'll see all of that show up. But part of just doing work now is you need a token budget. That that's the future of how we think about and how you budget for a team or a department. Mhmm.
Speaker 2:Where do you have any exciting case studies on the non software engineering agentic work side of the business? Because when agentic coding came to software development, there was a shift in a lot of software developers' workflows where they went from writing a lot of code to reading a lot of code, reviewing PRs. They they started operating at higher and higher levels of abstraction, but the decision what to build, how to prioritize things, how to build features, there's still conversations and software engineers are still doing work. What does that look like in the context of, of a finance department or operations department or a sales group? Like, where when the rubber meets the road, like, where are you seeing?
Speaker 2:Like, okay, this this is ready. This is the next, this is the next sub area or category that will really, really benefit from the productivity gains from agentic workflows?
Speaker 8:Yeah. I mean, a general thing, and then maybe a more specific thing. Yeah. What we're seeing is, like, you how do you measure productivity? Is itself a challenge?
Speaker 8:I mean, this is Yeah. This predates AI. Right? Like, lines of code, number of pull requests
Speaker 2:Yeah.
Speaker 8:Number of bugs fixed. The the I don't think I've ever heard a good engineering leader say that's how I measure the success of my team. Right? It shows up in other ways. Do you have a great product people love?
Speaker 2:Yep.
Speaker 8:Is your cost structure good? What's your install base look like? So so I think we have to get better at capturing that final business outcome because that's what's gonna get the CFO to open up the pocketbook Mhmm. And continue to invest Mhmm. In more tokens and more AI.
Speaker 8:So if I go then take that and apply to kinda other disciplines, like, you you what you don't wanna say is, oh, our salespeople send a lot more emails or our finance professionals create a 100 more spreadsheets. Like, that's more like of a nightmare scenario for AI.
Speaker 1:1,000,000,000 spreadsheets per company.
Speaker 2:No. No. I mean Exactly. Yeah. We saw someone on the timeline say that their their token budget went way up, but their revenue didn't go up.
Speaker 2:And so they're like Yes. I don't know if it's, if it's worth it. Like, you gotta show results.
Speaker 8:Exactly. So that that is what the question's gonna be everywhere. But the the good the good news is engineers historically have resisted clear measurement of impact because, you know, there's a little bit of art to it. But, like, sales professionals or finance professionals or customer service and and support reps, they are so metric driven. Like, if I'm a sales rep, what's my quota?
Speaker 8:What's my attainment? What's my revenue per head? Every single sales department has all of those things instrumented. Mhmm. And we've seen with m three sixty five Copilot, we kinda did a bunch of experiments.
Speaker 8:We published a white paper on it. Like, an almost 10% increase in revenue per seller just by giving them this the family of m three sixty five Copilot tools and things like our sales extension. Like, that's real dollars in the bank. Mhmm. Same thing for, like, back office finance processes.
Speaker 8:We've seen people use, like, our agents to do, like, accounts reconciliation, and we have, like, a great case study we did there where it was, 30% of all back office finance was automated out by this thing. So those those are examples of real top line and bottom line benefit. Mhmm. And that that's what everybody's gonna be talking about the next couple of years. It's it's not gonna be, you know, emails, documents, files, etcetera.
Speaker 2:Yeah. Yeah. I mean, you're in a you're in a high level business role yourself. I find myself doing a lot of like knowledge retrieval, deep research reports. But have you personally been using agentic?
Speaker 2:You mean this product specifically in your day to day? Like, even as something as simple as, a Sam Altman told us that he has a self, he has some sort of, like, to do list that does itself or something. I still don't understand exactly what that product is, but how are you using these tools these days?
Speaker 8:Yeah. So the for for, like, Copilot, Cobork, for example, I I've I've loved that. Like Yeah. My favorite magic moment has been around calendar management. Oh.
Speaker 3:I was
Speaker 8:telling this story to someone, like, I go go look out three months in the future and try to see, you know, how crazy is my calendar and travel look like.
Speaker 2:Interesting.
Speaker 8:And I was able to decline 17 meetings for me.
Speaker 2:Okay. That's I'd be recommending it. There. Yes. Exactly.
Speaker 2:Nothing gives me more joy than declining a meeting. You sold me.
Speaker 8:Yes. ASR. And I'd say, like, I went to Outlook and just saw all the meetings disappear and it was such a serene moment for me. And like, it did a deep analysis. Yeah.
Speaker 8:It it went and said like, who's my manager? Who do I work with? Yep. Am I optional or required? What's the topic?
Speaker 8:Yep. And it recommended the 17. And I went through all 17, said, you know what? Decline. And it was right.
Speaker 8:I mean That's great. So, that's one thing that I use it for. Like, another thing that I like right now is, I mean, I have an engineering background, so I'm not gonna go create the most beautiful content all the time. So Sure. I use it now to pull together information, design docs, meeting notes, and then go create either like architecture diagrams, presentations Oh, or spreadsheets or docs.
Speaker 2:Yeah.
Speaker 8:And I mean, I know it's just more doc generation, but it's very much more kinda like fire and forget. So it works really Yeah.
Speaker 2:Yeah. I mean, even if it's just like the bones of a of a PowerPoint presentation with the actual data pre populated that would just previously be a lot of copy paste, go run this analysis, copy it over, make sure it's the same font. Like, there's so much grunt work that Yes. Can be automated with these tools. I'm super excited for that.
Speaker 2:I I where do you see the the design PowerPoint generation tool use? I mean, people have been talking about, like, it's hard to create an RL environment with a verifiable reward around a great PDF, but it's easy to do that with coding. Do you do you have, like, an internal dataset of, like, the best PowerPoints ever created in your that you're, like, RL ing against? Or where do you see that going?
Speaker 8:Yeah. I mean, I think that is something that that there's a little bit of taste in all of this because like depending on the company, some companies like lots of infographics Yep. And lots of colors. Yep. Other companies like super fact based, no images.
Speaker 8:So like, you know, there's a little bit of an eye of the beholder, but absolutely, we have a a very robust harness in, like, gym environment where we're able to quickly evaluate models, prompts, tools in our own, like, fine tuning efforts to see if it's producing better content. So Yeah. I think, like, this like, a PowerPoints, Word documents, spreadsheets, it's gonna be like code. Yeah. And you you're starting to see it Mhmm.
Speaker 8:Already. But it's gonna be like code for sure.
Speaker 2:Yeah. I I mean, it's like it it a lot of these things are like XML under the hood anyway and and can be puppeteered like code anyway. Well, congratulations on the launch. Jordan,
Speaker 1:I hope you don't mind. I've been calling Copilot, co worker, Coco. Coco. So think It's very cute. Unless you like it.
Speaker 2:It's a very cute name.
Speaker 1:Yeah. But congrats congrats on the launch and great to Thank meet
Speaker 2:you so much for taking the time to come on the show.
Speaker 1:Cheers, Charles.
Speaker 2:We'll talk to you soon. Let me tell you about MongoDB. What's the only thing faster than the AI market? Your business on MongoDB. Don't just build AI, own the data platform that powers it.
Speaker 2:And let me also tell you about Graphite. Code review for the age of AI. Graphite helps teams on GitHub ship higher quality software faster. There's an interesting data point from Apollo featured by Restructuring here. AI is making starting a business easier and the numbers confirm it.
Speaker 2:Found Did you see this? Weekly business formation is exploding higher. So, previously
Speaker 1:matches what, the Collison brothers said when they did the show. Yeah. They're seeing not just more business formation but businesses growing Yeah. Faster
Speaker 2:It could be that AI is actually making businesses harder to start, but but Stripe Atlas is making it so much easier that we're seeing an upward trend, and it's it's fighting back against that. No. Of course. These are
Speaker 1:Weinstein is just so
Speaker 2:No. No. These are multiplicative very clearly, but Stripe Atlas does make it very easy to start a business. US business applications jumped insanely after the COVID pandemic. In 2019, we were seeing about 60,000, 70,000 new business applications, I think a four week moving average.
Speaker 2:We've been over 120 ks for years now, and AI seems to be driving that trend upwards by about 20 over the last two years, which is, great news. More more small businesses, they are extremely edifying, extremely valuable. I highly recommend it.
Speaker 1:We got another white pill.
Speaker 2:Give it to me.
Speaker 1:The data center build out has allowed Loudoun County to lower property taxes by 38 since 2010 Wow. Translating to average savings of about $3,400 per year for homeowners.
Speaker 2:This is great.
Speaker 1:Love it.
Speaker 2:What's happening here? Data center square footage in the millions has gone from 5,000,000 square feet to 40,000,000 square feet. That's a huge jump.
Speaker 9:Yes. This is like Northern Virginia. This is like AWS East one.
Speaker 2:Okay.
Speaker 9:Yeah.
Speaker 2:So so own homeowners are paying they went from 1.2% of whatever their house is. So if they have a million dollar house, they will be paying $12,000 a year. And now they're at point 8% property tax rate because they're taxing those data centers. This is the easiest solution to every AI problem. If the companies are profitable, tax them.
Speaker 2:If the if we already tax every business, we we we tax them by default. If somebody makes a lot of money on an AI stock, taxes is capital gains. You get the money. The money shall flow. And so good news for, the folks over in Loudoun County.
Speaker 1:Jacob says America once named things plainly, place trade surname. You had business international business machines, long distance telephone. Studebaker. Studebaker. And Bethlehem Steel.
Speaker 1:Steel.
Speaker 2:That International Business Machines logo is Goes hard. Is this real? Or is this like a reimagining of what the IBM logo could have been? Because I love this logo. What a very cool thing.
Speaker 1:That is apparently the original IBM logo.
Speaker 2:Wow. We don't know how to make logos like that in this country anymore, seriously. I mean, that that's what kicked all this off. Somebody was was upset about the granola rebate
Speaker 1:America's innovation slowdown is primarily due to the amount of Lord of the Rings that founders have to flip through before identifying name they can raise off of.
Speaker 2:No. You can just start a company and name it International Business Patrice. Whatever you wanna do, just name the company exactly that. Technology and Business Programming Network. It's that easy.
Speaker 2:In other news, Two Chainz.
Speaker 1:Two Chainz revealed he was an early investor in SpaceX.
Speaker 2:This is huge.
Speaker 1:He says, it just felt like I got in like early Bitcoin.
Speaker 2:There are some crazy early SpaceX investors. I don't know what Elon was doing with that early roadshow. We know some of these folks. Very cool community early SpaceX.
Speaker 1:Two chains.
Speaker 2:Get me two chains. Two chains got in at some point. And Gabriel over at OpenAI says, two chains is about to be three chains. Well, let me tell you about Okta. Okta helps you assign every AI agent a trusted identity so you get the power of AI without the risk.
Speaker 2:Secure every agent. Secure any agent. Let me also tell you about the New York Stock Exchange. Hopefully, we'll be seeing SpaceX there soon. Wanna change the world?
Speaker 2:Raise capital at the New York Stock Exchange. You gotta match two chains and Elon ringing the bell at the New York Stock Exchange. I can. I hope it happens. Well, we have our next guest in the newsroom, Beck from Sequoia Capital.
Speaker 2:He's partner there. Welcome to the show. What's happening? Julian, good to meet you.
Speaker 6:Hey, guys.
Speaker 2:How you doing?
Speaker 10:Nice to see you. We also hope to meet and ring the bell very soon.
Speaker 2:Holy holy Logomog. Yeah. Holy Logomog.
Speaker 1:Hear those companies.
Speaker 2:I do. We're we're we're partnered with Cisco. Thank you for backing them early, making TBPN possible
Speaker 1:Through your
Speaker 2:years ago.
Speaker 10:You just paved the way.
Speaker 2:Since this the first time in the show, why don't you give us just a brief backstory on yourself, how you wound up at Sequoia, what you focus on, what your day to day is like?
Speaker 10:Sure. So first, thank you both for having me. It's, it's an honor to to be there. I know you had a lot of my partners as well, so we're very excited to to be on the show. Yeah.
Speaker 10:My name is Julian. I'm a partner on the team. I'm based in London, but it's been a lot of time going back and forth to spend I'm with the rest of the team in California. And I started as a founder a very, very long time ago in high school. Wow.
Speaker 10:Built companies in hardware before.
Speaker 3:Cool.
Speaker 10:And so that's why I've been investing mostly in software Okay. For most of my career.
Speaker 1:Yeah. Wow. You you you were like, this burned out. I gotta get out of
Speaker 2:I was
Speaker 10:like, man, it's hard.
Speaker 2:You thinking about going back to hardware? Is software dead? What are you thinking?
Speaker 10:Yeah. Well, I mean, two things can be true at once. I think hardware is really interesting.
Speaker 6:Mhmm.
Speaker 10:I'm spending a lot of time thinking about physical AI. Sure. But at the same time, I've been investing for the last ten years in application software.
Speaker 2:Okay.
Speaker 10:So I wanna remain true to that, and this is kind of the things I've been spending time on recently.
Speaker 2:Yeah. Take me through your current thesis on how software is changing in the age of AI, in the age of maybe some sort of takeoff. AGI is here according to Sequoia Capital amongst many other people. The labs are recursively self improving. And it seems like Sam Altman made the claim that, like, if you want to build a startup, you shouldn't build a company that is in in the direct path while more and more products seem to be in the path of of AI and AGI?
Speaker 2:How are you thinking about the changing dynamic with software as we know it?
Speaker 10:Yeah. That that's a really important question, and I think we've been listening to Sam a lot as as shareholders.
Speaker 2:Yeah.
Speaker 10:And recently, just talking to our founders Mhmm. A lot of them are wondering if they're just an iteration away from the models replacing what they're doing. Yeah. And it it puts us in the position of thinking what is coming next, what what is going to be defensible. Yeah.
Speaker 10:And for us, we we've been making that prediction last week that the next trillion dollar company will be a software business that masquerades as a services firm.
Speaker 3:Mhmm.
Speaker 10:And we shared that thesis with with everyone online. Mhmm. The the reality is if you sell tools today, you're really in the line of sight for the models. And you're effectively competing with the next generation that they're they're going to launch. Whereas if you sell the work, you're actually benefiting from what the models are doing and all the billions of dollars that are going towards AI.
Speaker 10:So part of that prediction was saying for every dollar that's spent on software, $6 are spent on services. Sure. Until now, we could only really go after the $1 because we were building tools. Now that we can actually deliver outcomes, companies should think about capturing those $6 instead.
Speaker 2:So the, I mean, the obvious, like, first step is just get to outcome based pricing, move away from a seat based model probably. Would you agree with that as sort of like the most immediate change in business structure that needs to happen? Or if you're starting a new company, maybe don't start with seat based pricing on day one? Or is the, like, seat based pricing is dead? Maybe that's like an overblown characterization.
Speaker 2:How do you think about just, like, the business model moving forward?
Speaker 10:Yeah. We think both can work. They just serve different purposes.
Speaker 6:Okay.
Speaker 10:We have companies in our portfolio that start as these copilots Yeah. Right, that are helping the workers, and then we call them autopilots, the one who, you know, do everything end to end. Many of them are transitioning from copilot to autopilot Mhmm. To to do what you're describing. Mhmm.
Speaker 10:A company that comes to mind is a business called Sierra Yeah. That was started by Brett Taylor
Speaker 2:Yeah.
Speaker 10:That basically helps companies with customer support
Speaker 2:Mhmm.
Speaker 10:Which is pretty direct bull's eye use case of AI. Yeah. And in this example, they're actually charging the customers per outcome, right, per resolved ticket. And so if you spend $20 per ticket resolved with humans, they charge you, say, $5 instead. Right?
Speaker 10:And and that's driving ROI for the company. It's driving an outcome that's very measurable. And so that's the first more obvious company that's really working doing that. Frankly, until recently outside of software engineering, we haven't seen a lot of this sort of, you know, fully owned end to end outcome dry driving. Yeah.
Speaker 10:But the models are really progressing quickly, And for everyone who's been by coding the last couple of months, I think we've all been impressed that we can code a full website just end to end without any skills, frankly. Yeah. And and basically, prediction is that this is gonna happen to other categories. Mhmm. And we've been mapping mapping out these categories and helping founders come up with frameworks to prioritize where to attack them and how to do it.
Speaker 1:Yeah. Sorry to interrupt. One one thing I'd I'd I'd be curious to get your take on. So if a company is just selling enterprise software that helps people accomplish a task that feels very threatened by a foundation model company that maybe makes an agent that can also do said task and then you don't even need the software. So if a company transitions to just selling selling the work, selling the service like you're saying, you would agree that they're still gonna be under under threat from the labs who will also presumably be selling intelligence or agents that can go and do the work.
Speaker 1:But you're saying it will maybe be more closer to a fair fight. Is that is that the right way to think of it? Because I still think let let's say if somebody is like spinning up a instead of making a legal AI tool, they make AI native law firm and they're selling the work. I could still go to a ChatGPT, a Clawd, a Gemini and a future version of the model. Can just say like, hey, I need to do this thing.
Speaker 1:Can you help me do it? So there's still gonna be that competition. But you're saying that's a much better sort of competitive environment to be in than just selling like the software that enables somebody to do the task. Is that is that roughly right, or do you is is there something I'm missing?
Speaker 10:Yeah. I'll I'll just add one framework that I think is helpful for people to clarify what what that means. Basically, everyone is conflating the model's intelligence with what human are really good at, which is judgment. And I I think if you strip those two concepts that you really understand what the future will look like. And so right now, the models are becoming really good at everything intelligence related.
Speaker 10:So think of tasks that are, you know, rules based where just logic gets you to the goal. Right? Whereas the judgment is everything that humans really spike at, which is more harder to define. Right? It's instinct.
Speaker 10:People call it taste. They call it experience. And those things are things we think the humans will remain really good at for a very long time until intelligence is able to absorb these skills into it. But it's just really hard, right, because we we give the models a lot of credit for their intelligence, but they're just trained on set amounts of data. And so we actually think that this hybrid motion of coupling the intelligence from the model with the judgment from the humans is going to help you build the the next generation of companies, which we call these autopilots.
Speaker 10:Right? So true example. In the law firm, you will automate a lot of the intelligence work that's very much codified, which is what the model will do. But if you put the employee at the center, someone who's really experienced, can talk to a jury, who can, you know, mark a pause when at the at the right time, those are all the things that you learn only from real world experience that a top lawyer will be able to experiment, but that the models might will omit today. Does that make sense?
Speaker 1:Yeah. I still I still I like, you know, may maybe I'm maybe I'm too AGI pilled, but I but I just assume that a sufficiently advanced agent would be able
Speaker 2:to
Speaker 1:capture and accumulate, you know, the Two abilities to
Speaker 2:AGI pill. Just talked to Doctor. Alex Wissner-Gross for thirty minutes and he's like, my I'm uploading my brain right now.
Speaker 1:No. It's it's it's not that but it's just it's just the idea of like a, you know, you you put a smart person into an organization where they don't have any experience and they start doing things and they start getting feedback from other humans in the organization and over time, they accumulate enough knowledge and domain expertise that they're able to excel at something. Mhmm. That even even using even using the the customer service example, you know, you can you can you've always been able to outsource customer service and you can go work with a firm that is excellent at customer service. The alternative has always been to effectively hire a really great leader for that part of the organization, have them build out a team and a structure and an approach Mhmm.
Speaker 1:And execute Yeah. Something that that would be a similar result. There's like a cost trade off. But It
Speaker 2:does feel like overall there's like maybe some pressure on the b to b to b model or like the b to b to c model. If instead of, to use your example of of customer support, in, there was a time when when software companies would sell Phonetree software to, not to companies directly to do customer support, but to customer support outsourcing agencies that would then deploy that. And now you're seeing more companies go directly to, like like, the the the the firm that makes the the the the customer support experience end to end go directly to the company that needs customer support as opposed to having this, like, middle level. And it almost makes me feel like we're we're maybe going to rotate even a little bit further into d to c. When I think about law, there's a lot of stuff where the software would go to the lawyer, the law firm goes to the person.
Speaker 2:And if the person can go straight to the law firm or the the the legal model, you you wind up with a little bit more b to c activity, a little bit more company directly to to the the service provider or the technology provider. But but, again, this feels a little bit dependent on specific market dynamics. I'm interested to know your reaction to that, but also what does hold in your mind in the near term AGI scenario? Because there's been a lot of debate over how will marketplaces perform or are are network effects still super sticky in a world of AGI? Or what what are the what are the powers, the seven powers?
Speaker 1:Yeah. I guess I guess part of it The forces sticking out. My thoughts. Like, I I agree I agree with the entire thesis and like the relative Yeah. Short term like the one Yeah.
Speaker 1:One to two years. But then beyond that, there's still some uncertainty There is. Yeah. Like, I I don't think it Yeah. Solves the the sort of discount that even traditional SaaS companies are getting.
Speaker 1:Sure.
Speaker 2:Sure. Sure. Yeah. Julian?
Speaker 10:On on that point, I think you're you're right then. Basically, what we realized is that no one really knows. Right? I've been in many partner meetings recently where everyone just is astonished at how quickly things are moving.
Speaker 2:Yeah.
Speaker 10:I've been doing that for ten years, and it's just by far the fastest rate of change I've noticed. But some of my partners, you know, Alfred Lin, Pat Grady, they've been doing that for a lot longer, and they also acknowledge that everything is moving so quickly. And so what we're trying to figure out is also not what is going to change, but what are the things that will remain true no matter what. Right? Customers will always want things faster, better, cheaper.
Speaker 2:Sure.
Speaker 10:Right? And and so that's kind of the thesis that we're describing. They're just categories that are require more intelligence than judgment. And so we think that in the immediate, the ones that require more intelligence will will be actually really interesting for these autopilots because you can have you know, instead of having 10 humans in one AI, you have one human and 10 AIs. Right?
Speaker 10:And just that that ratio is just gonna shift as the the models get better and are able to absorb what we used to call judgment from the humans that are that is turning to to intelligence. And just to put, like, a practical framework to that, you mentioned outsourcing. It's it's actually, the best wedge right now is just to focus on outsourcing if you wanna capture the this services spend Mhmm. Because there's three reasons. The the first one is that outsourcing has already been agreed to a third party, so it's very easy to sell, right, because it's already a vendor that's external to the company.
Speaker 1:Yeah.
Speaker 10:The second is that there's an existing budget that's assigned to it, which you can peg yourself against. And and the third is that it it's actually very straightforward to swap a supplier. It's not so straightforward to reorganize your company. Right? We all talk about headcount reduction, you know, in terms of ROI, but actually putting that in practice takes time, whereas swapping a vendor is very quick.
Speaker 10:So I actually think that and that's the prediction is that the the companies that will succeed very well with this autopilot model will start with where there's a lot of labor spend that's being outsourced today. We mentioned customer support. We'll mention also things like accounting, tax, insurance, and and so on and so forth.
Speaker 2:Is there a world where in the next five years or something like venture capitalists truly shift away from software and go back into rare metal or something crazy like that. Like, I just keep I just keep going back to, like, the like, Daniel Gross, now at Meta, you know, venture capitalist. I think one of his first AGI trades, the first bet was, like, in the world of AGI, is copper mispriced? And it was. It doubled in price over two years.
Speaker 1:Yeah.
Speaker 2:And I'm just reflecting on this idea that, like, if the job of venture capitalist is to back software companies, that's one thing. But if it's if it's just to allocate capital in in a way that can have exponential upside and single investing could return a fund.
Speaker 1:Would it would it ever make sense to stop stop investing in software? Yeah. I would hate that because we obviously software. But it it has been funny. A lot of people have been pushing like, you know, the models are so good.
Speaker 1:We just need better harnesses. Yeah. And to me, this is giving me flashbacks Mhmm. To early GPT days Mhmm. Where people were like, GPT is really powerful.
Speaker 1:You just need the right harness Right. Know, wrapper. All these like wrappers Yeah. Yeah. That had explosive growth
Speaker 3:Yeah.
Speaker 1:Very momentarily Yeah. Until and and so I do
Speaker 2:Most AGI pilled person ever over here. I love it. But but but what what what do you think about, about, like, you know, the the the more physical aspects of of the job? Like, rare metal mining, for example. Does that does that seem like something people will be working on a couple of years?
Speaker 10:Yeah. I mean, every everyone's talking about the fact that energy will be, you know, the biggest bottleneck, right, to train these models. So it it's very natural that you would go there. Personally, that's not really where my skill set is. So I I really hope you're wrong and that we'll find other areas in between.
Speaker 10:I think the
Speaker 2:the the the real path will be somewhere in the middle.
Speaker 10:Hopefully. Right? Yes. In in my my my way of doing things has always been to focus on talent. Mhmm.
Speaker 10:And so right now, I'm just always asking the smartest people I know. Yeah. Who do you know? Who's the smartest person? What are they working on?
Speaker 10:And you're you're right. More and more people are going more into the physical world. Mhmm. But right now, that probably means things like robotics and things like that. People are making a lot of money trading commodities, but that that's a that's a different world.
Speaker 2:Yeah. Who knows? Maybe that's the job of the venture capitalist in 2030, just trading copper all day long. Exactly. You mentioned using AI to build a website despite maybe skills that have languished since your time in software or hardware.
Speaker 2:But how are you using AI tools these days? Are you still doing mostly like knowledge retrieval, market research for your job in venture capital? Or have you actually built tools that have accelerated certain things? Have you automated any of your workflows? I'm just kind of curious how people are using AI these days.
Speaker 10:Yeah. Well, I I I don't know. I hope my competitors are not listening. So
Speaker 2:They are. So you could say, I haven't touched it, actually. It's useless. Yeah. Yeah.
Speaker 2:Exactly. Exactly. You should uninstall organic
Speaker 1:Yeah. Organic capital
Speaker 2:Pen and paper, actually, is underrated. Maybe you should say Exactly.
Speaker 10:Back to the basics, pen and paper Yep. And really don't try cold code. It's Yep. It's really, really great. Now jokes aside, it's been really a transformation and Okay.
Speaker 10:Particularly the first quarter of this
Speaker 1:year. Mhmm.
Speaker 10:I I mean, I use it for all kinds. You know, the the the work is basically sourcing, picking, winning, company building, and harvesting. Right?
Speaker 1:Then Yeah.
Speaker 10:Yeah. I really have workflows that fit all these boxes. So it goes from things as, you know, at the end of the week, it will just, like, scan my entire calendar, look at all the meetings I took, look at the granola notes, which is a transcript Sure. Do an analysis and and help me, you know, try and find the key questions and the due diligence. Yeah.
Speaker 10:And, you know, it'll help me just prioritize. Right? Because this job is just so much about prioritizing. It's a power law. Right?
Speaker 10:There's only three meetings in a year that will really that could change your the course of your career. And and so making sure that you're attuned in those meetings is really important. Mhmm. So so that's really what I would say Cloud has helped me do right now. And I'm sure, you know, OpenAI will release something next month that will be amazing and so on.
Speaker 10:And you just have to experiment. Right? And I I think that's also, like, what we've seen with things like OpenClub being so top of mind is, like, this culture of experimentation is not just in software engineering now, but it's it's transpired across the entire knowledge workforce. And, you know, venture capitalists are not an exception.
Speaker 2:I love it. Well, thank you so much for taking the time to come shout out.
Speaker 1:Nice to meet you.
Speaker 2:Have a great rest of your day. Guys. We'll tell you work. Good bye.
Speaker 10:Doing that.
Speaker 9:Great day.
Speaker 2:Let me tell you about 11. Build intelligent real time conversational agents. Reimagine human technology interaction with eleven Labs. You know, I if I was a venture capitalist, Open Claw. Every morning at 5AM, text me the name and market cap of the company that I passed on three years I need inspiration to grind harder.
Speaker 2:That's the secret to success as is Lambda. Lambda is the super intelligence cloud, building AI super computers for training and inference that scale from one GPU to hundreds of thousands. And without further ado Up next. We have Eoghan McCabe, the CEO
Speaker 9:legend.com,
Speaker 2:a huge supporter of TBPN as well. Welcome to
Speaker 1:the show. How are you doing? Great to see you.
Speaker 3:To see you also It's on two
Speaker 1:long. Man, you have you just have a voice for for For
Speaker 2:radio. Livestreams, podcasts, all of
Speaker 3:my time.
Speaker 1:It's it's incredibly Yeah. Calming.
Speaker 2:I love it. I love it.
Speaker 3:Okay. What would you like me to say? I can record a meditation for us. We could Oh.
Speaker 8:Kinda start
Speaker 3:at the hour that way.
Speaker 2:Do you have time? Do you have to funding announcement.
Speaker 1:Do you have time to meditate these days?
Speaker 3:It's kind of the only thing that keeps me sane.
Speaker 2:That's good. Well, you have some new funding for new projects. Break us down. What happened? What what's the plan?
Speaker 2:How much did you raise? Take us
Speaker 1:through it.
Speaker 3:Yes. We raised $250,000,000.
Speaker 2:Let me hit the gong. Congratulations.
Speaker 8:Hit it. Did
Speaker 3:that gong get bigger since I was here?
Speaker 1:It did get bigger.
Speaker 8:It did
Speaker 1:get bigger.
Speaker 3:Would be cool if it got bigger every every single season. Right?
Speaker 2:Fast take off.
Speaker 3:Is that a thing?
Speaker 1:Just wait. There's a fast take off in
Speaker 2:There's fast take off in Gong size over here. It's growing exponentially.
Speaker 1:Except this was actually, I think, the largest we could find in The US. So we gotta go. Yeah.
Speaker 3:No. There's always bigger. I kinda have Gong envy right now. Wish I had a Gong.
Speaker 2:We'll send one over.
Speaker 3:Let's talk about that later.
Speaker 1:Well, we'll talk. Anyway We'll talk.
Speaker 2:So $250,000,000. What's the plan?
Speaker 3:Yeah. Yeah. So, I mean, we've been building this thing called Fin for three years now. Service was just this obvious place where AI was gonna, you know, blow up And all of sudden everyone else started to chase it. We were about a year ahead of everyone else.
Speaker 3:And we started doing, you know, customer queries and then more complex problems. Originally it was answering questions for customers and it's fantastic, you know. The typical response rate with a human was one to two days. You'd have to deal with the beautiful imperfections of humans. Unfortunately, these bots are just far more perfect than humans and they answer in seconds.
Speaker 3:And of course, they're cheaper too. So we started to say better, faster, cheaper. These things are really better, faster, cheaper. So that was a tremendous success. Finn is the biggest service agent in our category, nearly 100,000,000 ARR, 8,000 paying customers, you know, some of the biggest lowest in the game, thank thank you from Anthropic to
Speaker 2:Yeah.
Speaker 3:I don't know, Snowflake, the cool guys like Polymarket and Vant and everyone else are on fit. So wonderful. But where does it go next? You know, clearly it's just so obvious as each of these categories develop that there's just more to explore. None of these categories are in any way fully baked.
Speaker 3:Yeah. And that's one of the interesting things about kind of all of these AI spaces that we're still kind of pulling the thread and seeing what's going to show up next. And the next big play for Fin is not just answering service queries, but really getting deep into every stage in the customer lifecycle. And so we call that the customer agent. And this money will primarily bring the customer agent to market.
Speaker 2:Okay. So when I want great customer service for a company that I'm running, I want, yeah, you know, search the knowledge base, answer the obvious questions, handle the refunds, track. That's part of the customer journey. Are there other touch points that are unlocked by AI along that journey?
Speaker 3:Absolutely and completely. Yes. From day zero on the website when the customer has very basic questions, when they get more interesting, when the business wants to start to qualify and understand who that customer is, when the customer actually then wants their hand held when they sign up, or they want to actually engage and learn and meet salespeople, they want to book a demo, they want to talk to a virtual salesperson, I mean that's just the very top of the funnel alone. But to get into the application, if it's B2B, there's all the onboarding, all the detailed help required to get something successful. If it's e commerce, it's actually finding products and getting it into the cart and bringing them back for more shopping.
Speaker 3:So there is really the literally the full customer lifecycle of work to be done by agents. Mhmm. And customer service was just the little sliver. I wanna know what's happening off camera, though.
Speaker 2:Oh, we we just have our next guest walking in into the studio. Got it. Got it. Got it. So I'm just
Speaker 3:I'll figure they're less interesting than me.
Speaker 2:Yes. I I I I wanna talk about the some of the best customer service agents that I've hired have sort of wound up
Speaker 1:Real
Speaker 2:people. Real people have wound up in sales roles almost. Like, they have a fish on the line. They're reeling them in saying, you don't wanna cancel. We actually this product does exactly what you want.
Speaker 2:And Right. That the fact that it was delivered one day late, we're happy to refund you. But it seems like the underlying issue is that you actually want this product instead of this product. Why don't we get you over here? And that's this like sort of different skill set that comes out in the best customer service agents.
Speaker 2:How are you thinking about actually driving incremental sales through customer service agents?
Speaker 3:Yeah. I think one of the interesting things is that a lot of the work in these functions isn't entirely separate. Yeah. We just needed to draw lines between them when we were running human organizations because you can't possibly have one person who's outstanding at service, sales, marketing success, etcetera etcetera.
Speaker 1:Yep. Yeah.
Speaker 3:The SaaS market is too
Speaker 1:complicated. You know?
Speaker 2:They were
Speaker 1:all Totally. Yeah.
Speaker 3:A pain in the ass. But unfortunately, pain in the ass is coming with AI because I just don't believe you're going to have these separate sales agents and marketing agents. You're not going to have these things fighting with each other, having different context and information, different memory in different parts of the customer lifecycle, different messengers, not going to happen. There's going to be vendors that will provide one seamless solution across the customer lifecycle. And that's a big thing we're passionate about.
Speaker 3:And one of the very fascinating things we've seen, we've been billing this for quite a number of months now, and we're going to launch the first part of it next month. One of the fascinating things we've seen is that when there's an underlying relationship with the agent, and then there's actual connection in the moment, and then there's clear intent. The context is primed for selling and encouraging the user to do something next. Mhmm. If you're on a website and you've got a super annoying pop up that says, hey, John or Jordy.
Speaker 3:In fact, they're not even going use your name. They're going to say, do you want to learn more about our whatever? And you're like, get out of here. It's like not the right time or place. Whereas if you're actively discussing, it can ask you questions back, and it's so much more effective.
Speaker 3:And so we've seen just these things. The service agent, when it starts to do sales, for example, and vice versa, they're far more effective. And so the real magic from AI across the board, and of course this agent world, is gonna be when the agents can do things that humans can't.
Speaker 1:Yeah. One one that stands out is like right now you have this spikiness in dialogue as a customer of a business. Right? You might question coming in and it like creates a moment and then it goes away. And then you get maybe transition to someone else and you're talking to real human.
Speaker 1:You talk with them a little bit and you get passed to someone else and like, there's something about creating, I think the customer experience of having this continuous conversation with the organization that you're a customer of that I think will be super powerful because it just means that, you know, a split second away the entire time. I I I was talking with some guys that own a number of like big car dealerships recently. And I was talking to them about AI and they weren't fully sold yet because they they use like chat GBT but but they haven't had anything spread across their whole organization. And I was explaining to them how there's so many different moments where like a customer lands on their website and they wanna know if a car is available but somebody gets back to them too slowly and like they they the light was like finally going off to them. Do you think that there's a number how many industries out there have been kind of resistant to even using
Speaker 3:Mhmm.
Speaker 1:Software from Silicon Valley to date that are going to kind of take a second look at all the different tools and things like Fin. Now that they maybe have had a cool experience with AI in their Mhmm. In their personal life and they're basically saying they're they're willing to kind of look because maybe the the historical SaaS platforms, they tried them but it wasn't quite magical yet.
Speaker 3:Yeah. There there'll be so many spaces, so many verticals that some smart entrepreneurs will benefit from bringing sexy technology to, that aren't immediately obvious to those of us in Silicon Valley. You know, we tend to sell to ourselves. It's really insular. So all the names I just mentioned there, like Anthropic and Snowflake, they're all the cool brands in our world.
Speaker 3:However, there's, like, car dealerships that are, you know, I don't maybe it's not as big as as Anthropic, but they're pretty damn big
Speaker 2:and Depending on the car dealership, I I would put car dealership above Anthropic and Snowflake personally. If it's, like, a Ferrari dealership, Lamborghini dealership, like Fine. Just on the cool side, we gotta give it
Speaker 3:to the supercars dealers.
Speaker 2:On the
Speaker 3:aura factor, sure.
Speaker 6:For sure.
Speaker 3:Maybe not on on the economic factor. Yeah. But there are there are so many industries that just don't have our attention. And so there will be, you know, there'll be companies like ours selling to many verticals, but really, you know, we'll focus on all of the verticals that are adjacent to ours. Mhmm.
Speaker 3:And there's, you know, tens of billions of dollars of opportunity just in that alone. Yeah. But there'll be people who will say, we're gonna make agents for dentists, car dealerships, for certain government bodies, maybe even for tax bodies. Like, there'll be people who'll do this because so much of the effort is not just in building the technology, it's actually creating the trust, as you say, Jordy, with the actual people who will have to deploy this technology and make it work for them with their legacy systems. So there's a lot of very unsexy work that will be rich or rewarding for some people who bring it to these places.
Speaker 1:Talk about the decision to raise debt instead of equity. Not something you see Yeah.
Speaker 3:I mean yeah. I know. I I just realized I had this kind of blind spot for this as an option for for capital. And there's just so much equity opportunities out there. We raised a bunch last year for a secondary.
Speaker 3:But when you're in our dynamic where you're accelerating as quickly as we are, like every every year we we're doubling our our rate of growth pretty much. There's no way we're gonna be able to achieve our valuation that we're pretty confident we'll get next year.
Speaker 2:Yeah.
Speaker 3:And so any equity raise is gonna just be massively dilutive. We calculated that raising this in debt is gonna cost us about a tenth the kind of price to shareholders that it would cost in equity. Wow. So the the this is, you know, the incumbents have disadvantages. They tend to move a little slower.
Speaker 3:They have older businesses they need to wrangle with, but they have access to a nominal amount of debt. Like, 250,000,000 is a relatively small amount of debt for the size of our business. And I think that you're going to start to see a lot of these older SaaS companies as they pivot AI Mhmm. And they actually find new growth, start to lean on debt as an opportunity to invest rather than, you know, super dilutive equity. And honestly, I just think it takes a little bit of maturity to take that route.
Speaker 3:There's so
Speaker 1:much maturity. It's it's it's confidence
Speaker 2:Yeah. To know your business.
Speaker 3:Well, you know, I like to think so, but what I well, the reason I say maturity is that there's so much temptation to go and attain that latest, hottest valuation. Your employees want it, your previous investors want it, media wants it. Yeah. But if you're able to tell them we don't need it
Speaker 2:Yeah.
Speaker 3:Actually, it's way cheaper Yeah. For us to get a lot more capital with far less effort, then you can actually thread a needle in a way that I think is gonna work much better for big companies like ours.
Speaker 2:The gong over here rings just as loud for debt as it does for equity, so don't let them Yeah.
Speaker 3:I didn't actually You guys
Speaker 1:are hiring a ton. You said you're you're bringing out 650 new people Wow. This year?
Speaker 3:Yes. Yeah. So that's gross. I think net, it'll be a little less than that. It's a phenomenal amount of hiring.
Speaker 3:I fear it. I celebrate it with great caution. There are just countless stories of the hubris of CEOs who posted about how many people they've
Speaker 2:Yeah.
Speaker 3:Hired, and it's all ended in tears. So it's actually not a point of pride, frankly. Yeah. I think small is beautiful, but we are growing incredibly fast, and we need all the help we
Speaker 6:can get.
Speaker 2:And you've and you've been at this long enough to know the impact of what, you know, a 100 people, 600 people, what what impact that will have on your organization. So I Yeah. I have the full faith in you.
Speaker 3:Thank you
Speaker 2:so much for taking the time to come chat with us.
Speaker 1:Great to see you. Of course.
Speaker 3:Good to see you both. Well, care. Keep doing your thing. All of us.
Speaker 1:Amazing stuff. Goodbye.
Speaker 2:Let me tell you about Railway. Railway is the all in one intelligent cloud provider. Use your favorite agent to deploy apps, servers, databases, more while Railway automatically takes care of scaling, monitoring, and security. And let me also tell you about Sentry. Sentry shows developers what's broken and helps them fix it fast.
Speaker 2:That's why a 150,000 organizations use it to keep their apps working. And without further ado, we have Alex Epstein. He's the author of Fossil Future live in the TBPN UltraDumb. Welcome to the show. Good to see you.
Speaker 1:See
Speaker 2:you. Thank you so much for coming down on short notice.
Speaker 1:What are your what are your what are your texts been like the last forty eight hours?
Speaker 2:This is over. Because you're
Speaker 1:because John and I were thinking, who's our oil guy? We had one. We had one.
Speaker 6:Yeah. And I'm local too. So, yeah. If I'm in town, it's great. I was like coming in person versus
Speaker 2:the Thank you so much.
Speaker 6:Much. Other
Speaker 2:stuff. Yeah. Appreciate
Speaker 6:So, yeah, lots I mean, there's so much in energy, but I'm happy to let
Speaker 2:you lead the conversation however you think it should go. What are what's going on? What are what what how should we even be thinking about oil prices right now, the impact of the war in Iran? How do we how should we be processing this news?
Speaker 6:Okay. So we did this so there's a million things to comment on, so I'm gonna try to segment what I comment on.
Speaker 2:But
Speaker 6:we did get involved militarily in Iran Yes. Which I would just say in my non area of expertise, in general, that is a good idea. I think it's a better idea to have congress involved if you can get them involved, and I think it's a good idea to have a lot of expertise on international oil markets being brought to
Speaker 5:bear Sure.
Speaker 6:When you do this. And I think this administration does a lot of good stuff on energy, but I think it's pretty clear at this point that there was not maximum expertise brought in on this. And so the the standard issue when
Speaker 1:you're Indicators talking there would be like the strategic petroleum reserve hadn't been refilled.
Speaker 6:I mean, we we can talk about that. So let's let's make sure to get back to that one. Mhmm. But the main thing is just, you know, rough numbers, you know, straight of Hormuz, which Iran has control over, is 20% of the world's oil production is flowing through that every day. And so whether you're looking backward or looking forward, there is no replacement whatsoever for opening that straight and keeping it open.
Speaker 6:Mhmm. And one of the things we've heard from different people is, oh, we have a lot of options on the table. Like, there's a lot of things that we can do. There are definite things that you can do, but none of them is 20,000,000. Right?
Speaker 6:So it's just I mean, that's that's more than US oil production just flowing through that one place.
Speaker 10:Mhmm.
Speaker 6:So there yeah. We can talk about the other things, but I think the first thing people need to recognize is if you do not get that thing open, then you have dramatically higher oil prices. And then the other thing to get is oil prices really matter.
Speaker 2:From Muse. Oh, that's straight from Muse. Dumb question. Can't you just go around?
Speaker 6:I I mean, you can you can try to some extent, but there's a reason why 20,000,000 a day are being routed through.
Speaker 2:Because it actually comes out of there as well. Right?
Speaker 6:Yeah. There's just I mean, all of these things are very, very optimized. One of the thing about just understanding the energy industry is, like, things are just very optimized in terms of where the infrastructure is.
Speaker 2:Yep.
Speaker 6:You're doing all of these things. So it's there are other and we'll talk about this. I mean, there are certain pipelines you can pipe more oil through. There are alternate routes, but we're talking about in the millions a day maybe. Yeah.
Speaker 6:Most of our other options are sort of in the one to 2,000,000 barrels a day. Sure. And just so people know, barrels, 42 gallons, so you're talking about almost a 100,000,000 gallons a day for this stuff. But yeah, so so you don't have a lot of great options. And I think that's that's very important for I'm gonna give some options.
Speaker 2:Mhmm.
Speaker 6:But the number one option is keep that thing open. Mhmm. And there are basically two ways to do that. Mhmm. One is you just win the war, if you want to call it that, a lot more quickly.
Speaker 3:Sure.
Speaker 6:Like, if you get some form of surrender Yeah. And you have friendly people controlling it, then guess what? It can open.
Speaker 3:They'll
Speaker 6:open. That that's one thing. So I'm not an expert on how to do that, but schematically, doing that is very effective. Yeah. The other thing which is a little bit
Speaker 1:But but game game theory for Iran is they actually benefit from global markets being in turmoil because it gives them some leverage over over any type of negotiation?
Speaker 6:I mean, this is the biggest damage that they can do. I mean, clearly, we've seen they're not gonna do they're not gonna send missiles to Israel and get rid of Israel, let alone do anything to The US. Sure. But they have this incredible control of the one of the centers of the world economy. Yeah.
Speaker 6:And they have a bunch of stuff going on there. So let's talk about that. So option one is Mhmm. You get a surrender Mhmm. And they don't use all of their options.
Speaker 6:But then let's talk about their options. So they have mines Mhmm. In this thing. So they can blow up mines. Yeah.
Speaker 6:They have missiles. Mhmm. Maybe the worst thing they have, and this is a change in recent years, I know you've had Palmer on here and Ethan on here. Like, the drone thing is just a total game changer because you can have a small vehicle, you can launch one of these, you know, shaw heads out of the back of a truck. They can go really far.
Speaker 6:Yeah. And that's a lot harder to deal with than a fixed installation.
Speaker 1:Yeah. Have something like 500 miles of range on a typical
Speaker 6:Yeah. Depending on yeah. Right.
Speaker 2:So A truck drives up to a beach, drone launches, as soon as it finds that cargo tanker that has a bunch of oil on it, it hits it.
Speaker 6:Yeah. And then and then there's the you're you're worried about that
Speaker 3:Sure.
Speaker 6:As you're
Speaker 2:not gonna go.
Speaker 6:As as you're going through.
Speaker 2:So you
Speaker 6:have you have to think about those are the kinds of threats you face, so how do you deal with that? I mean, basically, need The US to lead some sort of convoy Mhmm. Where people have sufficient security and economic assurances that'll go through. So what's involved in that? Well, one thing is, if possible, you wanna get allies involved.
Speaker 6:And it's it's even possible you get unconventional allies like China Mhmm. Who has well, they depend on this. Okay. Now they have a bunch of unlike us, you wanna go back to SPR, they've been filling up their reserves. So they have more reserves relative to their imports Mhmm.
Speaker 6:Than we do relative to ours. If you think about ours, we have about 400 in the SPR, and we can release about 4,000,000 a day.
Speaker 2:A day.
Speaker 6:So we don't have as much but but still, they really care about this, and a protracted thing is bad. So but for sure, Japan, South Korea, India. So one thing is just at a high level if you're doing this convoy, can we get these countries that are aligned economically, maybe China, but definitely these others Sure. Can we get them involved? Because then it's an they have military presence there.
Speaker 6:That's an additional threat to Iran if they attack one of their ships. So but that's a macro thing is just can you do this alone or can you do it with allies? I think ideally you would do it with allies. Then there's a question of what you do about these these various things, and I think the biggest I mean, the other macro thing you can do is just in some way credibly threaten Iran and say, hey, if you and I can't give the details of how to do it, but if you attack us if you attack in the Strait Of Hormuz, it's gonna be very, very bad
Speaker 2:Mhmm.
Speaker 5:For you.
Speaker 6:So just at a high level, can you make that threat so they understand Credibly. This is gonna be very, very no. You know, we've wiped out a couple layers of leadership, so Yeah. They might be taking these things Mhmm. Seriously.
Speaker 6:Mhmm. After that, I think the biggest thing from what I've heard from military experts is just these drones. Like, what do you do about these drones? How do you give any kinds of assurances? And as far as I can tell, we don't have any one perfect solution, but you can do a certain amount of stuff from the air.
Speaker 6:Mhmm. You can do a certain amount of stuff from the ground, although it's very, very expensive. And then most cost effective where you can do it is you can take out certain stockpiles and facilities Mhmm. With the proviso that these things are decentralized. And of course, I have no specific knowledge about where they are.
Speaker 2:But in theory, if there's the suicide drone, the Shahid, that's there's a factory and you take out that factory
Speaker 6:Yeah.
Speaker 2:You have reduced that
Speaker 6:So that's gonna be it.
Speaker 1:And, you
Speaker 6:know, you you I know you guys had on Ethan Thornton, who I'm friends with, you know, he talks about this a lot. It's just this cost asymmetry issue, is you just wanna be doing things. Even if you're way wealthier, you do not wanna be spending $3,000,000 to take out $30,000.
Speaker 2:Which might be what happening with a Patriot missile battery.
Speaker 6:This this this. Yeah. That's the kind of thing that happens. So you need you need a combination Mhmm. Of those kinds of things.
Speaker 6:And then the other thing so that's all the security stuff. So you get the allies involved. You take these actions specifically against drones. I think the what was I gonna say about the the other thing you can do, I'm just making sure
Speaker 1:They're still just Well, like, other big
Speaker 6:thing Oh, the is insurance.
Speaker 1:Okay. Insurance.
Speaker 6:Yeah. The insurance. So and this this has been floated by the president, and it's it's a reasonable idea. But my understanding is the insurance vehicle we have, the development corporation, does not have enough funding, so you might need to go to congress. But, basically, you wanna be able to say, hey.
Speaker 6:You it's we're gonna lead this as The United States. We're gonna lead it with allies. Mhmm. If Iran does anything, it's big trouble for them, and we can counter their specific attacks.
Speaker 2:Mhmm.
Speaker 6:And if something goes wrong, you're insured. You're you're trying to create that confidence to get enough people going through, and then you get a certain amount of stability, and then you start to get closer to your 20,000,000 barrels a day. Either that or you just defeat them really quickly or both.
Speaker 1:So so the other factor is just the loss of production, disruption to production, plants shutting down, plants being damages damaged. How does that factor in? Do do those do those do refineries get brought back? And I'm talking about an allied allied country.
Speaker 6:You you mean just the the trend over time or what?
Speaker 1:Yeah. Generally, I mean just seeing seeing the videos coming out. You've seen places I think like Qatar, Kuwait, places like yeah. Basically just saying like, yeah, we're gonna pause production because we don't even have a place to store this
Speaker 6:Yeah.
Speaker 1:And or they're suffering damages and it doesn't make sense to keep, you know, refineries online.
Speaker 6:This is all but this is all just downstream of do you have the route open. Yeah.
Speaker 2:Got
Speaker 6:it. Like, if you have the route open
Speaker 2:Then you're good.
Speaker 6:It solves everything. Okay. If you don't, the the way to think of the other things, so we do have some other interesting options, we have to think of them as these are temporary stop gaps. None like, if anyone says I mean, you hear some crazy things. People are like, oh, Venezuela.
Speaker 6:Like, we talked about this last
Speaker 2:time. You.
Speaker 6:Venezuela is less than a million. It's just I mean, it's beyond crazy. It's just impossible. We it's not like Venezuela has the ability to just increase their oil production by any significant amount in the near future. Mean, we're talking about companies considering going in harm's way to get some incremental boosts.
Speaker 6:So Venezuela is essentially useless right
Speaker 2:now. So best case scenario in the next year, Venezuela goes from under a million to 2,000,000?
Speaker 6:No. I I mean Not even that. 2,000,000 would be insane. I mean, if we made it our entire goal in life, I don't know. But but why would you for that?
Speaker 2:Venezuela was at, like, 3,000,000, like
Speaker 6:Yeah. Yeah. Or something. Yeah. Some years ago.
Speaker 6:Yeah.
Speaker 1:Yeah. Yeah.
Speaker 6:Right. But but So 3,000,000.
Speaker 2:But there's a lot there. It's not gonna happen.
Speaker 6:It's just so yeah. So Venezuela, I would not even that's just in the category of not it's even a bad idea. It's just not an idea.
Speaker 3:Sure.
Speaker 6:There's nothing there's nothing there.
Speaker 2:No. No. That's right.
Speaker 6:So so the bad the bad ideas, by the way, are the worst idea imaginable is ban oil exports from The United States, which I have heard floating in conversation.
Speaker 2:Seems somewhat reasonable. I want
Speaker 6:the oil Of course. Because we have oil, so we don't wanna sell so, I mean, so many different things. But one obvious one not obvious one, but it's obvious if you know how these things work is the refiners in The US do not match up well to the oil produced in The United States. The the refineries in The United States are based on heavier crude. Like, crude is rated in terms of weight and sourness.
Speaker 2:Okay.
Speaker 6:And so our refineries are mostly for heavier crude. It's actually one reason why Venezuela has some appeal. Canada has a lot of appeal because they
Speaker 2:have heavier crude.
Speaker 6:Yeah. And we're gonna talk about Canada.
Speaker 2:We're good at refining it We're good produce a lighter crude
Speaker 6:Yes. Because the shale revolution was just this dramatic shift Yeah. Very quickly in oil production. So our refineries are not primarily equipped. That's why when we ended the crude export oil ban, I forget it was 2014 or something like that, it was just this huge unlock
Speaker 2:So
Speaker 6:because it allows the global mark it allows us to produce for a global market.
Speaker 2:You can't you can't look at American oil production as a monolith.
Speaker 6:Yeah. It's not it's not fungible.
Speaker 2:Import heavy crude and export light crude. Is that roughly correct?
Speaker 6:Roughly correct. Yeah. Yeah. So that's one thing is it's just insane Okay. To do that.
Speaker 6:But in general, think about it. What is the value of higher prices? Is it stimulates production? We're gonna talk about all these different ways where we wanna unlock oil.
Speaker 2:Yep.
Speaker 6:So you wanna tell American oil producers, hey. We're gonna actually totally screw you over. It's actually gonna be worse for you maybe than it was before because you don't even have a market. Yeah. So you're just strand all this oil.
Speaker 6:So that's that's terrible. There's also this financial manipulation idea, has been floated, which I think is terrible. Hey. Let's manipulate the financial markets basically in a way that we're you're selling futures short on a certain time period, and then you're buying them. What you're trying to do is lower the near term price.
Speaker 2:Oh, like Fed And raise
Speaker 6:the long policy. Or treasury. Yeah. So you're
Speaker 1:A very American way to try to solve Yes. So this world problem.
Speaker 2:A lot of people that like that.
Speaker 6:The only solutions involve unlocking oil Okay. In one way or another. You do not wanna screw up the markets Okay. So that people are less inclined to unlock oil. Mhmm.
Speaker 6:Nor do you wanna screw up the the markets to destroy information.
Speaker 2:Mhmm.
Speaker 6:So, again, the the main unlock is reopen that strait. Yeah. There's no way around that. But then we can talk about other things. So one one category is what you call spare or emergency capacity.
Speaker 2:Okay.
Speaker 6:And this this is there's some uncertainty here, but
Speaker 2:it Quickly. This is separate from the strategic reserve?
Speaker 6:Yeah. Yeah. So well, emergency capacity is the strategic reserve.
Speaker 2:Oh, it is. Okay.
Speaker 6:So spare means you could be you could pretty easily be producing more per day Mhmm. But you're not because you're not happy with the price. And this is mainly Saudi Arabia is is considered exhibit a in terms of amount of this. Now now there's debate among experts about how much they have. Yeah.
Speaker 6:But, you know, some people estimate they have one, two, 3,000,000 barrels a day of spare capacity. They could ramp up on some kind
Speaker 2:of And do they know how much capacity they have and they're just not sharing it, or does no one know?
Speaker 1:I mean, that's always been the story. They know. Been the story of The Gulf. Right? They're not gonna if they just flooded the market with every single barrel that they could possibly produce, they would be getting a worse a much worse price.
Speaker 6:Yeah. And there's this cartel arrangement that does. I mean, in a sense, everyone has this on different time scales. You think about US shale producers. If if if prices were sustained at a $100, guess what?
Speaker 6:There's a whole bunch of shale deposits that they could produce at $75 a barrel and make a fortune on that they're not gonna do
Speaker 2:Got it.
Speaker 6:Four weeks ago when it was around $60 Mhmm. A barrel. But but in terms of the thing about the Saudi oil is it's produced at a much lower cost. Okay. So this is them saying, hey.
Speaker 6:We have some of this on the table. And then the question is you need to be able to produce it Mhmm. And then you need to be able to transport it. So you have estimates around maybe you could get one to 2,000,000 barrels
Speaker 1:Mhmm.
Speaker 6:A day.
Speaker 2:Yeah.
Speaker 6:So that's one of them. Mhmm. In terms of emergency, if you look around the world, don't have a lot of immediate spare capacity besides that. So we have what's called emergency capacity or strategic petroleum reserve.
Speaker 2:Mhmm.
Speaker 6:You know, it's it's Jordi mentioned that we didn't fill up our strategic petroleum reserve. I think this is unequivocally, a mistake. I mean, if you look at the the whole thing of the strategic petroleum reserve is you want it for when you need it, and the best possible time to fill it up is when prices are low.
Speaker 2:And is that physically a place with a bunch of barrels that
Speaker 6:we just There's a bunch of different parts of
Speaker 2:of oil?
Speaker 6:Well, there's multiple different places. But but in you you can think of it But we basically cash
Speaker 2:the physical oil in America.
Speaker 6:Yes. Physical
Speaker 2:We own it
Speaker 6:or can access it.
Speaker 2:Yes. Exactly. And it's sitting there going unused until we're Yeah.
Speaker 6:Now there's issues of Biden. Biden definitely misused it, so they they used it to basically have lower gasoline prices during midterm elections.
Speaker 3:Okay.
Speaker 6:Even though parenthetically their entire policy goal was get rid of fossil fuels, which means you want the price to go up so people can't afford it. Yeah. Was a very cynical kind
Speaker 1:of move.
Speaker 2:Got
Speaker 6:it. They did it in a way that degraded the facilities. So let's say we're at about 4,000,000 barrels a day. We can get out of that thing. We have 400,000,000 in there.
Speaker 6:We should have over 700,000,000. We could have easily filled it up. Mean, it was a perfect time when you're talking about $50 a barrel oil, $60 a barrel oil.
Speaker 2:Yeah.
Speaker 6:Now it
Speaker 2:take And and and can I do the basic arithmetic of 4,000,000 a day, 400,000,000 in the reserve, a hundred days of oil? But but that's not actually if we're talking about, like, relieving price pressure, you could you could trickle it out over a year and have one twenty or one quarter of the effect.
Speaker 6:But just think about a $100,000,000 a day A 100,000,000 barrels a day is roughly the global market.
Speaker 2:Oh, okay. Okay.
Speaker 6:So so 4,000,000 is just that's its maximum Throughput. Capacity.
Speaker 2:Got it. Okay. It's kind
Speaker 6:of an analogy to batteries. Sure. I'm talking about batteries. So there's a certain amount like with a battery, if you hear there's a four gig there's a one gigawatt battery installation. That means that's the it usually means there's one gigawatt for four hours.
Speaker 6:Sure. But if you wanted to do half a gigawatt, then you could do eight hours. Yep. Right? Yep.
Speaker 6:It's it's the same deal with Okay. But keep in mind, this is not a hundred days of US oil demand. And by the way, for the refinery reasons, we can't just supply it all with
Speaker 2:our HPR.
Speaker 6:It's just the maximum output is but, you know, 4,000,000 barrels a day is something we're talking that's one fifth of what's flowing through the Strait Of Hormuz. If yeah. It's it's a But but, yeah, you can think of, yeah, maybe we'd be willing to do one to 2,000,000 barrels a day from ours. Now interestingly, the International Energy Agency we are part of the International Energy Agency reserve program. Mhmm.
Speaker 6:So we have about 400, but they have about one point four four hundred million. They have 1,400,000,000. So we have allies around the world
Speaker 2:Okay.
Speaker 6:That can also release this oil. So maybe we could get one to 2,000,000 from them. Now last at least last I checked the news, they hadn't agreed to do this because they don't think it's an emergency.
Speaker 7:Sure.
Speaker 6:But that's what it's there for. Yeah. So notice the pattern is we're
Speaker 1:At what point would they think it's an emergency? I don't know.
Speaker 6:I mean and we, you know, we can put pressure on them. I I think I think some of these determinations have to be made in conjunction with the military determinations. You can't make them in isolation. Because if you think about what's the military objective, how close are we to that? Or how close are we at least to opening this up via high security and financially backed convoys
Speaker 2:Mhmm.
Speaker 6:Then you can think about the timetable here. If you think about, well, the Strait Of Cornwall is gonna be closed for five years, let's just get used to being
Speaker 2:poor. Black bill.
Speaker 6:No. I mean, this is because people don't get oil. Like, oil is there's a reason why I focus a lot of my life on oil.
Speaker 4:Like Yeah.
Speaker 6:There is nothing more there's no material in the world of energy that is more valuable than oil because oil is just so unique in terms of it has this very high energy density and portability. And there there's nothing like I mean, nuclear has We're
Speaker 2:with plutonium.
Speaker 6:Yeah. Yeah. Doesn't have the same portability.
Speaker 2:We're gonna be rich. We're gonna have energy too cheap to meter in five years. So yeah. I talked to a lot of nuclear founders.
Speaker 6:Well, look. I love nuclear as much anyway. They don't they don't even have a portability solution, though, the near future unless you're talking about aircraft carrier or an icebreaker or something like that. So I'm just saying the world runs on mobility. Oh.
Speaker 6:And oil is ideal for mobility.
Speaker 2:So How portability in the sense of, like, using nuclear to actually power trade and commerce and moving ships around. Yeah. I mean Even if you can power the grid, you will be poor because you will not be able to trade. Is that
Speaker 6:what Yeah. Trade is really important.
Speaker 1:Trade is
Speaker 6:trade and well and also it's within the country too. We don't have the nuclear trucks to move things around. Sure. Sure. Sure.
Speaker 6:So you obviously want super cheap electricity, and you want super cheap transport And for and and we wanna see if we can electrify some of this stuff Yeah. If you can do it cost effectively. Yeah.
Speaker 2:There's some people working on, like, electric bulldozers, for example. It's very hard, and
Speaker 6:And and, yeah, by the way, guess what? You need a lot of oil to even get that whole supply chain started because you need the mine. Yep. So, anyway, bottom line, yes. I I meant that thing about we can all be poor very seriously.
Speaker 2:Yeah. Yeah. Yeah.
Speaker 6:Very seriously. So we have if you look at the spare capacity, think Mhmm. That's all of them. So we have the I the rest of the IEA. Mhmm.
Speaker 6:We have ours. Mhmm. And then to my knowledge, we have Saudi slash UAE. Mhmm. Them using spare capacity and then piping it where they can.
Speaker 6:So maybe combined, we're talking about 6,000,000 a day Mhmm. If you use all of those. I mean, somebody could imagine 10,000,000 a day. And all of these have time, particularly the emergency reserve things. These all like, the at least the Saudi one is you're sort of you have a deposit a big deposit here.
Speaker 6:It's just you have a small reserve, so you can't do this forever.
Speaker 2:Sure.
Speaker 6:And, of course, there's a risk of if you deplete the SPR more Yeah. What if you really, really need it Yep. In the future? And then there's a bunch of nearer term stuff that you can do. Mhmm.
Speaker 6:And some of which I like just because it could be done anyway. Mhmm. So number one to piss a whole bunch of people off is let's stop using this Jones Act. Do you know what this Jones Act thing is?
Speaker 2:I've heard of it, but explain.
Speaker 6:So the Jones Act is a set of restrictions that say that you can only transport things among ports in The United States if you have everything is basically American. So it's owned by an American. It's run, I don't know, all the it is all Yeah.
Speaker 1:It's basically American ship,
Speaker 2:American crew.
Speaker 8:Yes.
Speaker 6:Yes. So it's it's it leads to all these crazy crazy things like we end up importing things from like, instead of importing it from one part of the country to another, you end up importing from some really faraway place. It it leads to that kind of thing. In this case in particular, this this as Californians, we should all recognize this. Yeah.
Speaker 6:We don't have oil being piped into California.
Speaker 2:Yeah.
Speaker 6:So we want very efficient maritime transportation. And if you suspend the Jones Act, then you can get more efficient transportation, which means we can at least have lower prices here, and we can have less of a price shock.
Speaker 1:Mhmm.
Speaker 6:But in general, I want people to get used to having no Jones Act because I want there to be no Jones Act.
Speaker 2:Sure. Sure.
Speaker 6:So that's that's a good thing to do. Mhmm. My favorite idea here
Speaker 1:And is and the the I imagine there's a bunch of powerful lobbies that will fight and for the Jones Act because there's Correct. Entire industry built around
Speaker 6:Yeah. They say that only have a good shipbuilding industry. Like, I don't know what people think is happening with our shipping like, our ship building industry is not
Speaker 1:going as well as other people's. It didn't exactly work.
Speaker 6:Yeah. I mean, it worked to protect certain people's No.
Speaker 1:I'm just saying if you look at American shipbuilding, it's it's certainly
Speaker 6:Yes.
Speaker 1:Allowed us to be competitive on on a global stage.
Speaker 2:The shipbuilders would say we're early, but I take your point.
Speaker 1:Early, but you've seen but I'm just saying like
Speaker 6:It's like
Speaker 1:how many decades of declines
Speaker 6:Yeah. Okay. Well
Speaker 1:do you need?
Speaker 6:I I think John's like early.
Speaker 2:It's like a it's like a smiling curve like churn. It goes down and then goes back up.
Speaker 6:So let's let's That's
Speaker 2:the that's the bull case.
Speaker 6:That's that's the alright. Well, let's Yes. It's definitely not gonna help us. So so the here's the most interesting thing
Speaker 2:Yeah.
Speaker 6:Which I is a pet issue of mine Yeah. Is we're we're like, this administration, I should say, does a lot of good stuff on energy.
Speaker 2:Okay.
Speaker 6:I say that I you know, I'm in a fortunate position where I get a lot of people ask for advice and stuff like that.
Speaker 2:Mhmm.
Speaker 6:And I feel like one of my jobs is to tell people things they're not doing. Mhmm. A lot of things they'll just do right on their own. But but Canada is not one of those things right now. Mhmm.
Speaker 6:We are just absolutely sleeping on Canada as an opportunity. So Canada has no people and infinite resources Yeah. And is really friendly. Yeah. So you do they have these oil sands.
Speaker 6:You know what these things are?
Speaker 2:You know, explain.
Speaker 6:So so they're just like these deposits of oil. It's like nature basically committed an oil spill
Speaker 1:Yeah.
Speaker 6:Is the way to think of it. Okay. So it's all these oil sands. For years, they were not very useful.
Speaker 2:They weren't viable because it's not just you drill down, black gold comes up. It's like buried in the sands need to be
Speaker 6:They're kinds of things. You can mine it out directly or you can do what's called in situ, which is you heat it up underground.
Speaker 2:Oh, and it liquefies.
Speaker 6:In any case, we they have just this unbelievable oil seep. They have way more oil deposits than we do, way more oil reserves
Speaker 2:Okay.
Speaker 6:But they have on the order of a third of the production. Okay. And in part, because they have absolutely terrible policies, which are their fault. But in part, we have had terrible policies, including getting rid of the Keystone XL pipeline
Speaker 2:Oh, yeah.
Speaker 6:And this kind of thing. So but if you just think of Canada, it's just we should be so they have water. I mean, they get no people, one tenth the population, infinite raw materials. They got everything. They got timber.
Speaker 6:They got natural gas. Yeah. They got oil.
Speaker 2:Yeah.
Speaker 6:Like and we're talking about Venezuela. They're a friendly country. Right? Yeah. They got smart people there.
Speaker 6:They're they haven't been driven out of the country because they're afraid of getting killed by Chavez or
Speaker 2:Yeah. Yeah.
Speaker 6:Maduro or whatever. So this is an opportunity to say, hey. Canada, let's work. Let's have a task force. You guys are at a low in terms of we don't have pipelines, but we can do rail.
Speaker 6:At least we can do rail, maybe trucks, but at least rail. Rail transports from Canada are really low right now.
Speaker 2:Okay.
Speaker 6:So let's have an initiative where maybe you bring a couple 100,000 depending on your expectations, maybe we can add a couple 100,000 barrels a day. Mhmm. So that's one that excites me because I wanna be like, The US Canada superpower thing is this huge opportunity that we're sleeping on because of the idea of, oh, we should only be doing it in, quote, America. Oh, interesting. By the way, unbelievable.
Speaker 6:I think they have the best uranium in the world.
Speaker 2:Yeah.
Speaker 6:It's Maple syrups up there. Yeah. I don't have a Canadian personality, but I love that place.
Speaker 2:No. Strategically. Makes
Speaker 6:it And they're very friendly. Yes. They're very friendly.
Speaker 2:Maybe we should be friendlier.
Speaker 6:Yes. You know, we should, and we have trade agreements coming up. So there's that.
Speaker 2:Yes. Okay.
Speaker 6:The other thing is we have some you know, we can increase capacity here, as I mentioned, depending on prices going up. Yeah. So there's that. But that might be a couple $100,000 barrels a day. Mhmm.
Speaker 6:But so high level, open the Strait Of Hormuz. That's the only thing that really works.
Speaker 7:Mhmm.
Speaker 6:Then you can talk about Saudi Mideast spare capacity, our emergency capacity, and IEA emergency capacity. And then below that, you have Canada Jones Act and US That's great. Production.
Speaker 2:That's a really, really thorough list. Thank you. I appreciate that. Anything else, Jordy?
Speaker 1:What are you what are you gonna be paying attention to most closely?
Speaker 2:Do you watch the price chart? Is that important to you?
Speaker 6:Well, it's important. I mean, I I'm it's weird. I work with I run like a kind of a network of energy executives, but I'm focused on the policy side of things. Sure. So I'm often laughably out of touch.
Speaker 6:I mean, I know in general within $10 usually what it is. Yeah. But I as a policy, I don't I don't invest in energy.
Speaker 2:Looking at longer term five
Speaker 6:year But but also, like, I'm deliberately avoiding exposure to the commercial part of it because
Speaker 2:Of course. Yeah. You wanna be independent.
Speaker 5:Yeah. I would
Speaker 6:I wanna be independent. So That makes sense. Yeah. I'm not but at the moment, yes, I am looking I am looking at the prices. Okay.
Speaker 6:But but I'm I don't do predictions, but I'm just saying I think there's a service right now in just laying out the options for people because
Speaker 2:Yeah.
Speaker 6:Even in the government, I mean, you know, not everyone has a lot of expertise in Mhmm. In oil, and this is the kind of thing where, it's yeah, kind of why I think you texted me this morning. I mean, sometimes you need to know sometimes you need to know about oil today, and sometimes you don't. But when you need to know, you need to You need know me. You really need to know.
Speaker 2:Yeah. Well, it seems like whether you're in the business community or just, you know, humanitarian, everyone should be rooting for a swift and peaceful resolution.
Speaker 1:Yeah. Do you do you have any insight? This this might be totally out of your wheelhouse, but insight into how, how people running businesses that are most impacted by fluctuations in oil prices? Like, are they were they have they been actively hedging over the last couple weeks? Are they doing like, how how do you run your business when your input cost can go up dramatically potentially at a much higher rate than you can adjust your own prices?
Speaker 6:I mean, just in general, these guys are sophisticated, I think, within the bounds of doing it. I don't think you can hedge indefinitely for this big a price
Speaker 3:Yeah.
Speaker 6:Kind of differential like somebody is gonna say I mean but it's it's also you you think about there's on the consumer side, but on the production side is is even more dramatic if you just look at the margins of these guys. The I mean, the oil industry is just just fascinating. I mean, I just remember anecdotally when I I came out with my first major book, the moral case for fossil fuels in 2014. Even I could see speaking engagements correlated to these prices because you had a big price crash around them. Yeah.
Speaker 6:And then you think, wow. Like Yeah.
Speaker 1:We don't need we don't need Alex.
Speaker 6:Yeah. Variable yeah. It might be nice to keep the troops motivated, but, you know, they'd probably rather be employed.
Speaker 2:That's right. Yeah.
Speaker 6:But it's a hard it's a hard business. Yeah. And one of these things is you just the people who survive have a very strong constitution
Speaker 11:Mhmm.
Speaker 6:And and they they they can manage the risk very, very Mhmm. Well. I was talking with with a bunch of these these guys and actually some people we know in common. Yeah. And, like, some of the better ones, some of the two of the young superstars in in the industry, they're just saying, like, like, we love the chaos because we can just no.
Speaker 6:We can handle it, like, when the prices go down.
Speaker 2:Sure.
Speaker 6:Because that that's you think about what they've been dealing with over the last year. They've increased production. Prices have gone down. Yeah. And and you look at this.
Speaker 6:This administration is very focused, I would say, too much focused on having prices go low. Like, this is this is one of the variables here is although the administration, I don't think was fully was not fully prepared for this situation. In general, they are maniacal about prices going low. Like, Trump in his mind has a very strong focus on $50 a barrel, which I think he feels like that's always the perfect price for oil.
Speaker 2:Sure.
Speaker 6:And the people in the oil industry say, wait a second. Like, we've had inflation of everything else. Why does our one product
Speaker 2:Yeah. Yeah. Yeah.
Speaker 6:Have to stay at $50? And that's part of the appeal of Venezuela, rightly or wrongly, because it's not gonna make a big difference soon. But it's like how do we keep oil prices Low. Low. Yeah.
Speaker 6:But that's that's hard to be in that industry as a consumer
Speaker 2:Mhmm.
Speaker 6:As as a as a producer rather, and it's hard as a consumer. So they have financial instruments, but the main thing is you just you need to you need to just have the flexibility in your business in your business model. And it's it's oil is, again, it's the most valuable material in the world of energy Mhmm. Which is why everyone uses it, but it's also why it's so inelastic and you have these price fluctuations. So you just need to come to terms with that, or somebody needs to figure out something better.
Speaker 6:But it's it's really hard. And, of course, policy wise, we can have better policy to make it more stable, but it's still that thing where it's so valuable that guess what? You get a little extra demand, a little less supply, price goes up a lot. It's it's it's different than other things which you can substitute for much more rapidly.
Speaker 2:You're out of a job when we build the Dyson sphere.
Speaker 6:Oh, man. I got I'm I'm working on this AI thought leader thing we talked about last time. So I got Okay. I got plenty to do.
Speaker 2:We'll cross that bridge when
Speaker 6:we come I got plenty to do with my time. Yeah.
Speaker 2:You're Thank you. As
Speaker 1:long as we need energy, you'll you'll be doing us fine.
Speaker 3:I guess.
Speaker 2:Thank you so much for coming on the show. Let me tell you about Shopify. Shopify is the commerce platform that grows with your business and lets you sell in seconds online, in store, on mobile, on social, on marketplaces, and now with AI agents. And let me also tell you about AppLovin. Profitable advertising made easy with axon.ai.
Speaker 2:Get access to over 1,000,000,000 daily active users and grow your business today. And without further ado, we have our next in person guest live in the UltraDome, Michelle Volz. How are you doing? Good. Good to see you.
Speaker 2:Good to see you. Thank you so much for coming down to the TBPN UltraDome. Yeah. No. For those who aren't familiar with your venture capital career and your marathon running career, introduce yourself.
Speaker 11:Yeah. So I am Michelle. I was previously at Andreessen Horowitz. I just launched my own fund, Pax Yeah. Which is $501,000,000 dollar early stage venture fund focused on foundational categories.
Speaker 11:So I was on the American Dynamism team, Andreason Horowitz Yeah. Similar themes to that, just earlier stage.
Speaker 2:Biggest lessons from first. Oh, yes. We got him with the Gong.
Speaker 1:I think I think Michelle should do it herself. Okay. Here we go. Hit the gong. Here in person.
Speaker 2:The gong.
Speaker 1:Go for it.
Speaker 2:The mallet's already sort of broke.
Speaker 1:You won't break it.
Speaker 2:Smash that. Anyway, let's let let's start with like lessons from Andrew Steinhardt. It's like, what did you learn? What are your biggest takeaways? What are what's the playbook that you think you will be continuing?
Speaker 2:And then we go into what you think might be different?
Speaker 11:Yeah. I mean, I think in the categories that I look at, it is common for founders to need to raise a lot of capital.
Speaker 2:Okay.
Speaker 11:And the ability to be able to fundraise for that capital ahead of maybe what our traditional milestones in software investing is very
Speaker 7:important. Totally.
Speaker 11:You need to be able to be a magnet for capital, for talent, and sell a vision. Mhmm. Because what you're doing is like very hard with with fewer obvious proof points maybe at the early days.
Speaker 2:Yeah.
Speaker 11:And I think identifying that in founders early is something that Andreessen Horowitz is very good at.
Speaker 2:Yeah. How how how in in American dynamism companies, hard tech companies, it does feel like there's, like, high CapEx requirements, high human capital requirements. You're not gonna automate everyone on your supply chain on day one. And so there is this, like, how you talk about growth, how you talk about
Speaker 1:about
Speaker 2:scaling and success and finding product market fit, it's it almost feels back to like the, okay, we're scaling DAU and we're let's focus on that metric instead of like profit in software. What does it look like in hard tech? Is that the, okay, we have a program of record or we have these SBRs coming in or we have this long contract that is very solid, but we're not gonna draw down on it for a few years, so we'd have a capital need. But how how do you think founders in hard tech should be talking about what progress looks like at at top line or whatever top line means?
Speaker 11:Yeah. I mean, I think that's the the hardest part. Like, in software, there's such a structured set of metrics Yeah. Everybody has been trained around Sure. Like growth rate, like weekly users, revenue.
Speaker 11:It is so different company by company Mhmm. In these sectors and that's why the storytelling is important Mhmm. Because if you're talking to later stage investors, they might be trying to make a pattern recognition off of a pattern that doesn't quite exist yet.
Speaker 2:Mhmm.
Speaker 11:And so in defense companies, like, yes, there are government contracts you should be trying to get. You should be working towards programs of record, but there are like different intermediate proof points beef before that. I usually explain it to founders as like the three things you're trying to progress on are people, product, and then like revenue or traction. Mhmm. And so it's like, are you hiring the best people and the people that are experts in the different categories that you need?
Speaker 11:So you maybe you need a mechanical engineering hire, maybe you need a government sales hire, maybe you need software as well. Are you making progress on people? Are you making progress on product that's tangible? Like, can you show that there's like developments? Maybe that's like you've built something and it's like test like gotten to a test milestone.
Speaker 2:Mhmm.
Speaker 11:Maybe it's in the hands of users, like are you making product progress? And then, are you getting some sort of signs from customers that, like, they are willing to pay for this? Like, are you getting contracts? Are you getting LOIs? Like, are you getting something that shows that there's a commercial use case for this?
Speaker 11:I think sometimes in like hard tech, there's a million names. Hard tech, frontier tech, American dynamism. I think sometimes people skew more towards thinking it's just deep tech, which is like would project on this. Exactly. Like science risk.
Speaker 11:Like fusion is probably like the most extreme example. Yeah. That's hard. Like it's like, of course, there's a customer if you can do this, but like, we'll see what the timeline is hopefully soon. But I think in a lot of the categories I look at, there are commercial proof points you can make.
Speaker 11:Like, there's usually a latent demand or, like, or an immediate demand. It's like, how do you show that you can get those customers and they can, like, signal that they will buy from you?
Speaker 2:Yeah. There was this viral Suttrini article that said
Speaker 11:Oh, yeah.
Speaker 2:Software is dead. Sell everything. How are LPs thinking about hard tech as a category with that backdrop of, like, the SaaS pocalypse? I imagine it makes things easier for you, but has it actually rippled through the LP community to the degree that your investors are more excited than ever to have a hedge against chaos in automatable software, all of that?
Speaker 11:Yes. I I will say this with love to all of my LPs. I think founders are typically like leading indicators Mhmm. In categories. Like, they they are at the cutting edge, like, they are seeing around corners.
Speaker 11:Often, like, great founders will see something before VCs do. Like VCs are lagging indicators.
Speaker 2:Mhmm.
Speaker 11:Like once something is hype y, it's like usually too late to a good early stage investment. LPs are lagging lagging indicators Sure. Once something is very mainstream and there starts to be some signs of returns, then they're like, we should pay attention to this category.
Speaker 2:Got it.
Speaker 11:And so I would say, a thing that was more surprising, partly because I live in my own echo chamber and constantly hear about, like, all of these great companies building in, like, defense tech, hard tech, and, like, industrial tech. LPs weren't as familiar as I expected. So, like, there was a level of education. I think there's still some weariness of like, there haven't been a ton of exits yet. Yeah.
Speaker 11:I mean, I think that might change if SpaceX IPOs Yep. Will be one of the biggest IPOs ever in history. Yeah. But I would say, like, the the appetite for it grew over time as I was fundraising.
Speaker 2:On SpaceX, yeah. I mean, great point. It's it could be, like, 1,750,000,000,000.00 is the number that's being floated. It's it's staggering amount. It's gonna return, like, not just, like, one venture capital fund.
Speaker 2:It's gonna return, like, 20 venture capital funds. Like, everyone who invested at early stage is gonna see their fund return from that. But all the LPs had to go on a twenty year journey with, a million continuation vehicles. How are you thinking about timeline messaging to LPs? Because ten to twelve years is what I've heard recently is kind of standard.
Speaker 2:But I've talked to some VCs and say, yeah, like most of my LPs want their money back in like six to seven years. And yet in hard tech, it feels like it might be longer. Are you communicating a particular timeline or is that even a relevant conversation?
Speaker 11:Yeah. I think like every every fund roughly pitches ten to twelve years
Speaker 2:Sure.
Speaker 11:Sort of the expectation.
Speaker 2:Yeah.
Speaker 11:I I think it's actually an interesting catch 22 because if you had ownership in SpaceX, your LP is actually like, like, hold on to that.
Speaker 2:Yeah. Totally.
Speaker 11:Like, keep that. I think it's gonna keep going up. But maybe you would have an option to take some off the table Sure. And return some, you know, money to LPs if that's what they wanted. But often, if it's like a really really good deal, LPs are like, oh, I think keep going up.
Speaker 11:But I I think it's like case by case. I actually just was talking to LPs about this and they were like, sometimes it makes sense and sometimes like hold on and Mhmm. I think if you build a lot of trust with your LP is like they're usually along the ride with you.
Speaker 1:What what geos are you spending the most time right now? Obviously, the it feels like we have so many so many founders that are just like in in our kind of neck of the woods here, but are you spending a bunch of time in Texas, Bay Area? Where where do you expect to deploy most of your dollars?
Speaker 11:Yeah. I think, like, there's five geos. Mhmm. It's San Francisco, LA, Austin, New York, and DC. I'm actually seeing, like, a significant amount of companies in DC.
Speaker 11:DC. Austin, I would say, though, is the biggest surprise for me. I know people have been trying to make Austin happen for a while, and I I was skeptical. And now, I think, like, it's sort of been an explosion of great companies.
Speaker 2:It was kind of the same thing with LA. Like, there was this long LA movement a while ago. Snapchat was sort of, like, the first big, like, you know, public tech company or one of them to come out of LA. And then the El Segundo thing just happened, like, completely organically without any of those people that were set cheering for LA, and it just sort of popped up. And it seems like Austin's going through a similar thing.
Speaker 2:What's unique about the companies that are building in DC? Are they thinking more demand generation first, more more work on Capitol Hill first, and then maybe they'll set up a satellite manufacturing office or work with, a contract manufacturer to deliver their capability?
Speaker 11:Yeah. I think, like, you're seeing founders who are working at companies, like like Palantir, for example, that have a big office in DC. Yep. They've been relocated to DC. They're close to their customers Sure.
Speaker 11:And then they spin out, start a company. Usually, these companies will end up opening an office in New York as well. Mhmm. It's just a deeper pool for engineers. Mhmm.
Speaker 11:But there's a surprising amount of really strong engineering talent in DC that is pretty fixed. Like, they wanna stay in DC. They have a life. They have a house. They have a family.
Speaker 11:Mhmm. So it it can be a very good place to start a company.
Speaker 2:What's the view on shipbuilding? Jordy says it's impossible to build a ship in this country. He was saying it can't be done. Is it possible? Possible?
Speaker 11:We we need to figure it out. Otherwise, we're gonna be behind. I think Yeah. I mean, like, the optimistic take would be, like, we can make it possible. Sure.
Speaker 11:We should make it possible. The pessimistic take is if we don't make any changes into the regulatory environment or the cost of building these things Yeah. It might not be, and that would be a
Speaker 5:shame.
Speaker 2:Yeah. That makes sense.
Speaker 1:Strategy with the fund? Yeah. There are these, like, $2,000,000 lead Seed three seed type checks. What what's the deal?
Speaker 11:Yeah. On average, 1 to $2,000,000 checks, like early stage, try to be in like the pre seed seed occasionally a little bit later rounds. Like to partner really closely with founders. And I and I think the special thing about the early stage, which is just my favorite stage of company, is you get to be very aligned. It's like a significant check for me if I put, you know, 1 to $2,000,000 into your company.
Speaker 11:And I'm very motivated to help you be successful to get to the next round. And then for that next round, I can be much more strategic in in helping you get there because I can't lead your next round. Whereas I think sometimes at the multi stage funds, it can be a little bit tricky to give advice for future fundraisers because you might be able to participate. And so, you have to be like more delicate about it at a big fund and as a small fund, you just get to, like, be totally aligned with the founder.
Speaker 2:One last question about sort of the the category targeting. American dynamism made a lot of sense as, a theme around defense tech and hard tech, and you put a manufacturing company in there. One of the interesting subcategories of AD that I always that always popped out to me was something around education, like reeducating workforce education, also just, like, education technology broadly. Do you think that that's within your purview or is that something that you're sort of carving out and focusing more on the deep tech, hard tech, physical products?
Speaker 11:Yeah. I would say I like to look at things that sit at the infrastructure layer of society Mhmm. Which could include education Okay. And housing, sometimes even healthcare. Often things at the intersection of hardware and software or tech and government Mhmm.
Speaker 11:Or just like doing something fundamentally important for like the broad majority of Americans. I think American dynamism is this beautiful phrase that can be an umbrella phrase as well. Yeah. I I sort of embrace that as well with packs. But at the early stage, like, what you're really betting on is people.
Speaker 11:Like, there is so little to understand except, like, the team in a dream. Like, are you building something in a category that's very exciting? Are you taking the big swing? Do you sort of have a right to win in that category and some sort of unfair advantage? I like founders that have like worked in the industries that they are building in before.
Speaker 11:And do I believe you can be that magnet for capital and talent and everything?
Speaker 2:Yeah. Where where are the big talent pools these days? There's the SpaceX mafia, the Palantir mafia, feels like there's an Andoril mafia now.
Speaker 1:Base power base power maybe in Austin. Austin eventually.
Speaker 11:Yeah. I think is starting to see a few things. Basically, company that's that's like reaching escape velocity or like like escape fit. And product market fit has grown really quickly Yep. Hired exceptional people.
Speaker 11:There are some people that really love that zero to one stage Sure. And the company gets big. Like, think we're seeing it at Anderl. We saw it at Palantir. We continue to see it at Palantir.
Speaker 11:People people learn how to grow quickly and then wanna apply that to their own company.
Speaker 2:Amazing. Well, thank you so much for time.
Speaker 11:Thank you.
Speaker 2:I'm sure we'll have rest of your day.
Speaker 1:Have a lot of your companies on.
Speaker 2:Let me tell you about Vanta, automate compliance and security. Vanta is the leading AI trust management platform. And let's go over to some hard tech that's happening with Stargate. Because By
Speaker 1:the way, John Yes. Apparently, Trump said something to the effect of, I think the Iran war is very complete, pretty much.
Speaker 2:Oh, okay. Well
Speaker 1:So we may be on the verge of new era. Yeah.
Speaker 9:Crude is back down to like pre like Friday, pre weekend
Speaker 2:What does that mean? Prices? $80.90? 83. 80 Okay.
Speaker 2:Well, maybe he was watching. Maybe he was watching Alex Epstein, and Alex said, hey. We gotta we we we gotta wrap this up. We gotta stop. The price is too too darn high, says Alex Epstein.
Speaker 2:Oracle and OpenAI allegedly dropped their Stargate expansion lease, Meta swooped in the same day. Ben Pouladine says that's not weakness. That's a hot real estate market. And then he goes on to comment that the real story is OpenAI passed on six additional buildings because power won't be ready for a year. The power is is is the gating factor here, not the chips, interestingly.
Speaker 2:We are in the pow the power bottleneck, the energy bottleneck. And by then, Rubin replaces Blackwell. So why are you gonna build a Blackwell data center when you could build a Rubin data center? So why expand with last gen chips? The eight building, 400,000 GPU base build continues.
Speaker 2:This is smart capital allocation, he says. And so this is based on reporting from the information about Oracle and OpenAI sort of adjusting the plan for for Stargate. I am particularly interested in digging into the the Stargate of China. They have a project. That's the Chinese Stargate funded to the tune of 36,000,000,000, not not up at half 1,000,000,000,000, but still
Speaker 1:That's a very big number.
Speaker 2:It's a big number and it's growing. And I've been I've been digging into it a little bit, trying to learn as much as I can. We'll do a deep dive probably with someone at semi analysis who knows a lot more
Speaker 1:Yeah.
Speaker 2:About that.
Speaker 1:Chad, GBT is back at the top
Speaker 2:No way.
Speaker 1:Of the App Store leaderboard.
Speaker 2:You know, we do have to We
Speaker 1:Adam, GBT says GPT 5.4
Speaker 2:Sentences for gods. You know what that meme is from? Big Chungus? It's a good one. Anyway
Speaker 1:Yeah. So so I'm I'm kinda surprised by this for the the one reason being
Speaker 3:Yeah.
Speaker 1:I expected Anthropic's been spending a ton of money on marketing. They started ramping it up.
Speaker 2:Yeah. Yeah. I see that
Speaker 1:all over I expected them to basically be willing to spend whatever it took to stay on top.
Speaker 2:Yeah. But I mean, it's capital war. So both sides are gonna spend a lot of money. Yeah. Right?
Speaker 2:And, you know, there was a what? What's up?
Speaker 9:Wasn't the rise in in the Claude, like, ranking just because of the the DOW news and now that's, like, not on the
Speaker 2:So so yeah. That's what I was getting at. It was, when the Super Bowl ad dropped and we had some harsh words for the the nature of the characterization around ads, our favorite thing in Claude, led us to launching Claude with ads. But our question was, will this Super Bowl ad move the needle for the consumer app? Like, a Super Bowl ad is a massive campaign.
Speaker 2:Millions, tens of millions of people will see this. It went viral. Like, it's will it work on people? Will people see this ad that's, like, sort of deep in the weeds around how ads work in LLMs? Like, will this be enough for people to go and hit the download button?
Speaker 2:There's a hurdle there. Will they do it? And they did it a little bit. It went I think Claude went from, like, like, twenty sixth to eighteenth or it started climbing then. And then, you're right, the the Department of War narrative really, really shot up.
Speaker 2:But as with any story in tech, it has to go back and forth and back and forth. And the memo that came out on last Wednesday probably sort of reset the narrative there and and and changed how people were thinking about that. And also, these charts are based on momentum and so it it it's very, very hard to
Speaker 1:say Yeah. Thought had gotten to something like a million downloads a day according to Mike Krieger. Oh, yeah. And at the same time, ChatGPT last month, I think, did, like, 55,000,000 downloads. It's still not outpacing Yeah.
Speaker 1:Sure.
Speaker 2:Sure. ChatGPT last month. Paramount plus is number four in the free apps. Although, how can it be free when you have to subscribe? But it's sort of
Speaker 1:Yeah. That's a whole thing.
Speaker 2:But bullish for the Allison
Speaker 1:team. Krish says, u m f's ruined agencies so bad that I'm gonna have to start using saying gumption.
Speaker 2:Gumption. Gumption is really good. I like gumption.
Speaker 1:Hi. Hire for gumption.
Speaker 2:AI gumption. AI agents.
Speaker 1:We only hire for gumption
Speaker 2:noun version of gumption. Don't know.
Speaker 1:A friend of the show, Amjad, said not having coding experience is becoming an advantage. And Shurya says not knowing how to cook is becoming an advantage, CEO of McDonald's.
Speaker 2:I love it. Amjad has some great takes though.
Speaker 1:This is just Yeah. This is really funny.
Speaker 2:Easy riff. It's it's all in it's all in good fun. What else is going on? Lots of debate over this of the over the singularity pricing as Will Dupuy put it because Barely AI, which I believe is Trung fan, says cursor's internal analysis shows how hard Anthropic is subsidizing Claude code. Last year, a $200 monthly subscription could use $2,000 in compute.
Speaker 2:Now the same $200 a monthly monthly plan can consume 5,000 in compute. There it does feel like there's a capital war. I'm very interested to dig into the history of of capital wars in in winner take all markets, but also in oligopoly markets. We will You should write
Speaker 1:the book on capital wars.
Speaker 2:Capital wars are fascinating. I lived through it during the Uber and and Lyft era. There was there was a flippening for a little bit where Lyft was was at the top of the charts and
Speaker 1:was The friendly the friendly rideshare company.
Speaker 2:It was very friendly. They the the cars would pull up with a pink mustache on them and had a very very fun whimsical branding. And you would sit in the front seat of the car because you were I friends
Speaker 1:I see Claude
Speaker 2:Did you not notice?
Speaker 1:Pink mustache.
Speaker 2:It was a real thing. Yeah. So what's Lyft
Speaker 9:supposed to do?
Speaker 2:You're So when Lyft so when Lyft launched, people see Lyft and Uber as, like, commodities at the at the time. But Lyft was, like, revolutionary, and people were super black belt on on Uber because Uber at the time was only black cars. So they didn't allow anyone to just hop on the network. Like, it was not a three sided marketplace or whatever, a two sided marketplace. Like, you had to have a, like, a livery to TCV number.
Speaker 2:I forget what it's called. But, like, you had to be registered as a limo driver. And you had a Escalade. And normally, someone would call you and say, hey. There's a wedding, and we want you to come shuttle people around at this wedding.
Speaker 2:And they would pay you for that. And Uber went to those folks, and they said, hey. You are registered to legally drive this car as a professional driver. We want to book you through our app. And so the black car became Uber's main product, their only product.
Speaker 2:And then Lyft figured out through some interesting, you know, strategy how to allow people to just show up with a Prius or a Toyota Camry or Corolla or whatever. And so drivers on launch day, I remember, like, the launch week in San Francisco, would just show up in just some random car. But Lyft gave them all pink mustaches, huge pink mustaches. The team can pull up a picture of the huge pink mustaches on the front of the Lyft cars that were very whimsical, and and it was remarkably cheap because Uber had been somewhat subsidizing the black cars. They hadn't been taking a huge margin.
Speaker 2:They were losing money. But the Lyft cars were way cheaper because someone was just showing up with, like, their Toyota Corolla or whatever with their pink mustache on it. And so so you would hop in the front seat because this driver was your friend. There you go. The pink mustache on top of the what is that?
Speaker 2:A CR V or something? And so and so you would hop in the front seat. You'd they would ask you to pound the driver. Give him give him the knuckles.
Speaker 1:Give knucks?
Speaker 2:Give him knucks. This is the thing.
Speaker 1:No way.
Speaker 2:I'm I'm almost sure. So you'd be like, what's up? And then you'd get in the front seat and they would drive you. Of course, if you had a group, you'd be in the back seat. But then
Speaker 1:Everyone had to give Nux or just
Speaker 2:I I think the Nux might have been optional. But encouraged. Recommended. Because it was like, hey, we're buddies. And we're and we're fighting the war on the Borg.
Speaker 2:We're going up against
Speaker 1:But you were encouraged to get in the front seat?
Speaker 2:Front seat. Front seat in the in the Lyft. In the Lyft, when you're taking a Lyft, your friendly lift driver shows up, you sit in the front seat, you give them ducks. This is the thing. Pink mustache.
Speaker 1:Chungus That's crazy.
Speaker 2:Big chunk of sense of
Speaker 1:cards. Alright.
Speaker 2:So
Speaker 1:You you gotta hire a car, but you have to be absolute boys You do. Driver.
Speaker 2:But but truth be told, like a lot of the early lift drivers were, like, super cool tech forward San Francisco people. And so, like, you'd have a lot to talk about because, like, people were, like, they they they it wasn't professionalized at all yet. It was very much, like, like, two sided marketplace. You'd make a lot of money as a driver. Like, there was no margin compression yet.
Speaker 2:It was like free money everywhere for everyone. And it was a capital war, and Lyft raised a ton of money and gave and came into the market with a disruptively cheap product and and wound up delivering a really fun experience that was very, very cheap. And then Uber sort of fast followed and launched Uber what's it called? The normal one. Not black.
Speaker 2:UberX. UberX. UberX. Yes. UberX.
Speaker 2:And UberX was bring your own car. And then and then the rest of the game played out. And it became a capital fight. And for a while, like, the meta was, oh, you're you're you're you need a lit you need a you need a ride. You open Lyft and you check the price, and then you open Uber until you check the price.
Speaker 2:And people would check both prices constantly. And then and then over time, like, the liquidity and and power loss stuff came into it.
Speaker 1:John, do you remember Uber helicopter?
Speaker 2:Yes. That was the with Blade? I think it was a partnership with Blade which was the Uber for helicopters and it was a thing but in in like, you know, for a few weeks or something.
Speaker 6:Did you
Speaker 3:ever do it?
Speaker 2:No. No. Because Uber Uber's had a whole bunch of different like, stunty products like Uber, you know, Uber Yachts or Uber they did Uber ice cream where you could get ice cream delivered. That was like the precursor to Uber Eats. They did an ice cream delivery truck where they all they they they every Uber driver checked in with the with the field office, got a cooler in the back with a bunch of dry ice and a bunch of ice cream, and you could press a button in for, two dollars.
Speaker 2:They would bring you an ice cream. They even did, I think, Uber kittens, and you could
Speaker 1:ask Uber to bring Uber dogs or something?
Speaker 2:You could, like dogs or something? I remember that. Yeah. It was very funny. Anyway, lots of stunts, lots of ways to get attention, go viral, and that was sort of the the launch strategy.
Speaker 2:They would do that for whenever they launched new markets. They would come in, and they'd be like, we need to go viral on the local news. We need to let the town know that we've arrived. And so lots of lots of back and forth in the capital wars of twenty twenty three, twenty twenty four. It was fun times.
Speaker 2:I lived. I fought through them. What else is going on?
Speaker 1:I think that's a good place to call it.
Speaker 2:Well, thank you so much for listening. We will see you tomorrow at 11AM sharp. Leave us five stars on Apple Podcasts and Spotify. Sign up for our newsletter, tbpn.com, and have a great
Speaker 1:rest your an honor and a privilege.
Speaker 2:It has.
Speaker 1:And I hope by the time the show goes live tomorrow, the war is over.
Speaker 2:Me too. Me too.
Speaker 1:That would be I would love that. That'd be quite nice. Thank you. Thank you. Goodbye.
Speaker 1:Cheers.