Health:Further

Retail sales fell 0.6% in July. Inflation held steady at 3.4%. Jobless claims jumped to 209,000. Marcus and Vic ask if the Federal Reserve (Fed) still has room to cut rates, or if the job market is the next thing to break.

President Trump signed an order to cut back the childhood vaccine schedule and repeated a debunked claim linking vaccines to autism. Medical groups pushed back hard. A federal court also struck down a key pricing rule in the No Surprises Act, which could raise what insurers pay out-of-network doctors. Employers are dropping the three biggest pharmacy benefit managers (PBMs), and small businesses are bracing for 14% premium hikes in 2027.

Biotech kept raising big money. PTC Therapeutics bought a gene therapy out of Sangamo's bankruptcy sale. Epicrispr raised $90 million. Braveheart Bio went public in a deal worth close to $440 million. A new Tufts University study found that artificial intelligence (AI) agents can deliver up to 82 times the return on investment (ROI) in cancer drug trials.

Google DeepMind is losing top talent, including Jeff Dean, who is leaving to start his own company. OpenAI is restructuring ahead of a planned public offering. Tether passed its first clean audit from KPMG, and the Securities and Exchange Commission (SEC) pushed back its crypto rulemaking plan again.

Marcus and Vic break down what matters and what's just noise.

00:00 Intro

📊 THE ECONOMY 
00:57 Retail Sales Fall 0.6% in July | WSJ ↳ https://www.wsj.com/economy/weak-retail-sales-number-adds-to-softening-economic-data-c047dce9?mod=rss_socialeconomyfeed 02:04 Inflation Holds at 3.4% | WSJ ↳ https://www.wsj.com/economy/central-banking/inflation-federal-reserve-interest-rates-296d1430?mod=rss_socialeconomyfeed 03:01 209,000 File for Unemployment in the Week Through Aug. 8 | WSJ ↳ https://www.wsj.com/economy/jobs/u-s-jobless-claims-rose-again-last-week-29384997?mod=rss_socialeconomyfeed

💰 VC & BIOTECH DEALS 
04:21 Epicrispr Raises $90M for Gene Therapy | Fierce Biotech ↳ https://www.fiercebiotech.com/biotech/epicrispr-bags-90m-rewrite-treatment-rare-muscle-disease 05:10 Vaderis Lands $152M for AKT Inhibitor | Fierce Biotech ↳ https://www.fiercebiotech.com/biotech/vaderis-lands-152m-squeeze-akt-inhibitor-through-phase-3-and-regulatory-approval 06:09 Khartis Launches with $95M Seed | Fierce Biotech ↳ https://www.fiercebiotech.com/biotech/pfizer-chemistry-whiz-debuts-khartis-therapeutics-sights-set-thyroid-eye-disease

⚖️ POLICY 
07:17 Trump Signs Vaccine Schedule Order | WSJ ↳ https://www.wsj.com/health/healthcare/trump-signs-order-to-reduce-childhood-vaccines-repeats-unfounded-autism-link-e9c9651d?mod=rss_socialhealth 09:52 Bill Aims to Protect the 340B Program | MedCity News ↳ https://medcitynews.com/2026/08/340b-drugs-senate/

🏥 HEALTH PAYORS 
14:09 Insurers Seek 14% Premium Hikes for 2027 | Healthcare Dive ↳ https://www.healthcaredive.com/news/small-business-health-insurance-premiums-2027-kff-rate-filings/827247/

🏨 HEALTH SYSTEMS 
16:37 Healthcare Faces a Cybersecurity Crisis | Healthcare Dive ↳ https://www.healthcaredive.com/news/healthcare-cybersecurity-crisis-def-con/827428/ 18:41 HHS Moves to Decertify an Organ Network | Healthcare Dive ↳ https://www.healthcaredive.com/news/hhs-decertify-organ-procurement-group-patient-safety-failures-network-for-hope/827334/

💊 PHARMA & NOVEL THERAPIES 
22:19 PTC Wins Sangamo Gene Therapy in Bankruptcy Auction | MedCity News ↳ https://medcitynews.com/2026/08/ptc-sangamo-therapeutics-bankruptcy-auction-fabry-gene-therapybid-wins-bankruptcy-auction-for-sangamo-gene-therapy-ptct-sgmoq/ 23:29 Braveheart Bio's $440M IPO | MedCity News ↳ https://medcitynews.com/2026/08/braveheart-bio-ipo-obstructive-hcm-myosin-inhibitor-cardiomyopathy-brve/

🩺 HEALTH & WELLNESS 
24:39 Health Nerds Feed Data to AI Chatbots | WSJ ↳ https://www.wsj.com/tech/ai/the-datamaxxers-feeding-their-every-health-move-to-ai-df0af9b5?mod=rss_Technology

🪙 WEB3 
28:58 Tether Passes First Clean Audit | The Block ↳ https://www.theblock.co/news/business/2026-08-13-tether-kpmg-first-full-audit-usdt-clean-opinion-411760 32:48 SEC Delays Crypto Rules Again | CoinDesk ↳ https://www.coindesk.com/policy/2026/08/13/sec-cancels-long-awaited-proposal-of-reg-crypto-postponing-meeting-without-new-date

🤖 THE AI TRACKER 
34:58 Google DeepMind Loses Jeff Dean | The Verge ↳ https://www.theverge.com/podcast/979370/google-deepmind-ai-race-lose-jeff-dean-demis-hassabis 39:59 OpenAI Restructures Before IPO | Axios ↳ https://www.axios.com/2026/08/14/openai-executive-greg-brockman-ipo 42:21 Gemini vs. GPT: The Speed Race | Decrypt ↳ https://decrypt.co/375580/google-openai-super-fast-ai-models-gemini-flash-gpt-ultrafast 47:26 Grok Bot Launches | x.ai ↳ https://x.ai/news/introducing-grok-bot

-----------------
more at healthfurther.com

What is Health:Further?

Every week, healthcare VCs and Jumpstart Health Investors co-founders Vic Gatto and Marcus Whitney review and unpack the happenings in US Healthcare, finance, technology and policy. With a firm belief that our healthcare system is doomed without entrepreneurship, they work through the mud to find the jewels, highlight headwinds and tailwinds, and bring on the smartest guests to fill in the gaps.

Vic: If you enjoy this content, please
take a moment to rate and review it.

Your feedback will greatly impact
our ability to reach more people.

Thank you.

Marcus: Hey man, how's it going?

It's

Vic: going well.

Hot, hot here in Nashville.

Marcus: I, I'm sorry, I just cannot hear
you say that it's hot here in Nashville

after- … spending a week in Phoenix.

I just, I just can't hear it.

Yeah.

I mean, I've been here, I'm walking
around, I'm enjoying myself.

I've got long- Yeah … pants on.

Vic: Yeah.

Marcus: It's like, dude, this
is, this is not, this is not hot.

Vic: Yeah.

I don't hang out in Phoenix, so yeah.

Marcus: Yeah, you're smart.

You're smarter than me.

Yeah, yeah.

You're smarter than me.

Uh, all right, man, lots
of stories to dig into.

You ready?

Vic: Yeah, yeah.

Marcus: All right, let's dig in.

Uh, so we had a bunch of numbers
that rolled out this week.

Um, the first, we've got, uh, retail
numbers, and they came in quite

soft, uh, point six percent down.

The economists had projected
only point one percent.

Um, I don't think there's
anything shocking about that.

I mean, you know, we're-- we go
to the grocery store, we go to the

gas pump, we go to restaurants.

I mean, it, it's hard.

It's hard right now, man.

Vic: Yeah, that's right.

I mean, I think the Main Street
economy, which leads to what people

buy, is, is difficult right now.

I mean, the only bright spot is
if you own stocks, they, they're

up, you know, continuously.

But, but that's not the real economy,
and it's not surprising retail's weak.

People aren't spending money.

Marcus: No, no.

So I th-- I think this is
probably not great news for the

Republicans, um, in August, right?

So close to the midterms- Yeah … you
would-- I think you'd want a slightly

stronger economic, uh, signal than this.

Um, so, so we will see, we will see.

Yeah.

We are-- We're, we're gearing up, buddy.

No question on that.

Gearing up.

Right.

Gearing up for the midterm sea-season.

Uh, also, uh, we had the CPI come
in, and it does come in as expected.

Um, but it was not such a big
bump that it looks like Warsh is

going to move to, uh, raise rates.

Um, given how soft the economy is, given
how jobs are coming in, given how retail

is looking, it, it seems like he's got
a real needle for thread here between

trying to manage inflation, and since
inflation came in as expected, you know,

that seems like, okay, job is okay there.

But seems like he needs to strengthen
the economy, so he's probably not

in a position to raise rates, and he
needs to kinda just hold where he is.

Do you agree with that?

Vic: Yeah.

I think that's right.

I mean, to me, the, the-- if you're
looking at the two risks, like is

there gonna be a lot of inflation or
really struggling with loss of jobs?

I'm more concerned with the
jobs than I am inflation.

But, but I don't see him cutting
either, so I think that's right.

He's not, he's just gonna hold.

Marcus: And then jobless claims rose
again last week, so the number of

people who filed for unemployment was
two hundred and nine thousand, uh,

in the week through August eighth.

This, uh, this number continues to rise,
and it is moving higher than expected.

Um, some of the previous months have
been revised, uh, to the negative.

And so this idea that, uh,
I mean, I'm, I'm having

conversations all the time, Vic.

I literally just was, was talking to
a very, very talented person, um, who

worked in finance, uh, is, is here
in Nashville right now and has been

quite shocked at how difficult it is
for her to find her next position.

Um, and this is just not
an uncommon story, right?

Hearing it more and more and more.

Really talented, well-credentialed,
educated, experienced people

with really strong references,
great resumes, can't get jobs.

Vic: Yeah.

Yeah, I mean, I think, I
mean, we cover it every week.

AI is changing so quickly,
and there's, there's at least

uncertainty in the economic outlook.

And so I think on balance
companies are just holding.

Let's just wait and see what happens.

I'm not, I'm not sure that AI's
destroying jobs, but, but it's certainly

not making it where the people need--
they need more and more workers.

So we're in this kind of
limbo period, I think.

Marcus: All right, let's
quickly move into VC deals.

Epic CRISPR raises ninety million to
rewrite treatment of rare muscle disease.

It's a Series C backed by Sanofi Ventures.

Um, this is another biotech,
uh, another biotech project.

So, um, biotech is Fully, fully back.

Uh, large, large rounds, uh, rolling out
and, um, this is, this is another one.

Vic: Yeah.

And this is, uh, you know, in that
classification of a rare disease.

You know, I don't think there's
really a cure right now.

And so it's, it's great to see.

It's also a lot of money being
invested in, in what is really

important f-if you have this
disease, but it's not a huge market.

So I think just another-- I'm sort of
following you that capital is flowing into

biotech right now on all fronts really.

Marcus: Yeah, for sure.

Videris lands a hundred and
fifty-two million for a phase three

study in another rare disease,
a rare blood vessel disorder.

You know, biotech really finds its
home in these rare diseases, right?

Um- Yeah.

You know, you don't wanna try to
totally upset an existing workflow

and distribution channel for, uh, you
know, a, a, a highly available disease.

So these rare diseases, um, they,
they get e-exemptions from the FDA.

Uh, there's, there's a lot of
sort of patient advocacy and

support of them, and they can
support sort of novel go-to-market

commercialization pricing models.

And so that's why there's so many biotech,
um, uh, assets in this particular space.

Um, but again, hundred and fifty-two
million dollar series B, uh,

propelling this, uh, Swiss biotech's
inhibitor through phase three.

And, uh, just a lot of
capital in the space.

You and I don't invest in this
space, but it's important to take

note that it's, it's definitely
capturing a lot of the capital.

Vic: Yeah.

Marcus: Uh, and then the third story comes
from a former chemistry whiz from Pfizer.

I think this is Chartis Therapeutics,
but it starts with a K instead of a C.

Um, so I'll just say Chartis.

Ninety-five million for
thyroid eye disease pill.

This is a guy who has sold the company
to, to Gilead and, uh, you know, winners,

uh, attract capital pretty easily.

This is a Series B. It was led by Forge
Life Science Partners with Long, Longwood

Fund and Alexandria Venture Investments.

Um, but there were existing
investors, Foresight Capital,

Lilly Asia, and Next Tech Invest.

The goal is to push for this
potential pill for thyroid eye

disease into clinical trials.

Um, so again, more money going
into, uh, you know, into therapies.

Vic: Yeah, and this is another case
where there's an existing treatment,

but it's IV delivered, IV infusion.

Certainly, if they can move it into
a pill form factor, it's better for

the patients, better margins, longer
shelf life, easier to, to distribute.

All those things are much better.

Harder to do, but, um, that's the goal.

Marcus: Exactly.

Uh, okay.

Moving into policy.

We have got to start with
President Trump's executive order.

President Trump signs an order
to reduce childhood vaccines and,

uh, continues to talk about the
connection between vaccines and autism.

Obviously, this is highly contested,
highly controversial, um, but- This

is obviously also an important part
of, uh, RFK Jr.'s, uh, posture around

vaccines and the need to continue to
claw back, um, their use in, in society.

Obviously, it's, it's also happening at
the same time that we're having rises

of, um, uh, diseases like measles, uh,
that are, that are going around, and now

we're-- we, we have this executive order.

It's, it's unclear exactly what the impact
is going to be of this executive order.

Also, it's, it's-- it seems slightly
unusual because typically executive

orders are reserved to, um, move
in a direction where the president

already doesn't have authority,
but he has HHS, and clearly RFK Jr.

Right.

Is, is, uh, you know, is, is
aligned with this strategy.

So to me, this feels mostly like
marketing, branding, communications,

as opposed to a true policy
that couldn't have already been

implemented through the agency.

But what's your take on it?

Vic: Yeah, I think that's right.

Uh, I think-- I mean, no one's surprised
that Trump wants fewer vaccines.

He's been talking about that.

This is his second executive
order in that direction.

And I agree, I think it's marketing or
it's him trying to push Kennedy and HHS

to move more quickly than they are I
don't know that it's possible to move

that much more quickly, but, um, this
executive order, you know, directs

Kennedy to form a, a, you know, a task
force again, another task force, the next

ninety days come up with recommendations.

I feel like we've been-- we've heard
this story, and there's been several task

force, uh, but they'll do another one.

Marcus: You know, a little bit it feels
like, um, you know, RFK Jr. had that,

uh, that interview with Dana Bash and,
and that, that made the rounds across

news circles and, and, and social media.

And a little bit this feels like
Trump kinda backing him up, right?

Sorta saying, "Hey, you know,
that guy doesn't stand alone.

I'm here with him." Yeah.

"And here's, you know, and here's an
executive order." So this fe-- this felt

like a, like a vote of approval for RFK
Jr., um, and, uh, standing behind his guy.

Vic: Yeah.

I think standing behind and
also we're not slowing down.

I mean, even though people are gonna
be complaining or, you know, the, the

interviews are difficult, they're, they're
going full steam ahead in this direction.

Marcus: All right, this is an interesting
story, a new development, and shows

that maybe the lobbyists for the
hospitals are making some progress.

Bipartisan lawmakers are fighting
HHS's Three Forty B rebate push.

Um, this is from MedCity News.

A bipartisan group of six senators
introduced a bill to reform

the Three Forty B drug program,
locking in hospitals' use of

contract pharmacies and adding new
transparency and compliance rules.

Legislation would also kill HHS's
contested rebate pilot within

a year and replace it with the
National Data Clearinghouse

to catch duplicate discounts.

So if we remember this whole saga,
you know, HHS has their, um, has

their, their bill that they--
Uh, not their bill, their, their

program, I think it's called Noble.

Um, uh, and basically it's like
a, it's like a alternative to

Three Forty B, definitely not
advantageous for, for the hospitals.

The, the hospital sued.

Um, they were denied in the courts,
and so it looks like it's now been

elevated up to, uh, the legislature,
and we have a, a bipartisan group of

senators that are taking this bill in.

Um, I, I, I think this really underscores
just how important Three Forty B is to

the ongoing viability of health systems.

Uh, the fact that this constant drumbeat
of a battle, back and forth battle,

um, on, on Capitol Hill really, really
shows you that, that th-this matter, uh,

is i-in some cases a matter of life or
death for, for health systems, right?

And particularly for nonprofit ones,
particularly for those that are serving

the underserved and the growing population
of, um, un-underinsured or uninsured,

uh, people in the United States.

Um, without these, with all the other hits
that are coming in, um, a lot of these

hospitals are simply not gonna make it.

Vic: Yeah, I think that's right.

I mean, two, two observations.

I like this, this story,
this-- I like this movement.

I would really… I mean, I wanna
support the Congress actually trying

to create bipartisan legislation.

I think this is a pretty good bill.

It's adding a whole bunch more complexity
to the 340B system, which I don't love,

but, but the idea that three Republicans,
three Democrats get together and try to

create a workable solution is really good.

I'd like to see it get some momentum.

Marcus: Yeah.

For sure.

Vic: And it would be
good for health systems.

I mean, even getting them a few
years reprieve would be good.

Marcus: Li-listen, I'm, I'm all for
addressing, you know, some of the

inefficiencies or the v-very intentional,
um, deceleration of, of innovation that,

that, that the healthcare industry broadly
and certainly the providers have, have

been implementing for, for decades now.

But we, we have to do it in a
way that's not going to totally

destroy the system, right?

Yeah.

And, and I mean, w- I've been looking
at the confluence of, of change,

um, and I just think it's a It's a,
it, it's, it's a nightmare scenario

that we're walking into right now.

Something's gotta give.

Something's gotta give.

I, I would probably be just fine
with 340B being held off for at least

some number of years while we, while
we see what the, what the impact

of supplemental payments, right?

Uh, what the impact of
the new work requirements.

It, it's like there's a lot of
change that hasn't fully set in yet.

Vic: Right.

Marcus: You know?

Yeah.

And adding this 340B, y-you know, thing
on top of it, it just feels, it just

feels like one too many headwinds.

Vic: Yeah.

I mean, the healthcare industry is a
pretty big ship that they're trying

to fix, you know, as they're flying
along, and there's real people in these

systems that need to be cared for.

And so I'm in favor of, of making
change, but, but doing it in a, in a

measured, kind of step-by-step way,
not, not trying to do everything in

2026 because it seems like a lot, and
there's a lot of systems that are gonna

have trouble changing quickly enough.

So I don't know the details of this,
but I, I like the fact that it's

bipartisan and Congress is trying to get
its act together to, to pass something

to make it clear they should do.

Marcus: Yeah.

All right.

And then bad news.

Bad news for you and I, Vic.

We run a small business.

Vic: Yes.

Marcus: And, uh, the Kaiser f- Family
Foundation has come out and said their

analysis says that insurers wanna hike
small business premiums by 14% next year.

How many years in a row now has the,
the insurance hike been about 14%?

Vic: I mean, many, many years.

I, I don't know.

It's, it's, uh, maybe six or seven years.

It's been a long time.

And, and it, it's-- that's probably
four times the inflation rate, right?

So you are not gonna increase
your revenue at that rate.

There's no chance you're
gonna increase it that way.

So it just can't continue.

And so what happens is the employers cut
the coverage, but they're, they're not

gonna be able to keep the same coverages,
so they'll increase the deductible.

They'll add in co-insurance that
unfortunately no one understands,

and it, it is-- it means people
are gonna not have access.

End of the day, you're cutting access.

Marcus: Yeah, it's pretty insane, man.

I mean- Yeah … you know, it's, it's,
it's like clearly this is going to flush

out a bunch… You know, a-and, and I, I
think to some degree we're, we're hearing

from in-insurers, "Hey, uh, uh-" In some
cases, we'd rather just not have you.

Vic: Right.

Marcus: Right?

Like, like it's, it's, it's too
expensive to keep you as a customer.

Um, so if you can afford
it, you can afford it.

Uh, and if you can't, I
don't know, go do ICHRA.

Go do, y- you know, whatever.

But again, as we, as we have discussed,
as the ACA marketplace keep, continues to

thin out, ICHRA is becoming less and less-
Right … of a viable alternative, right?

So.

Vic: Yeah, and the risk
pools are, are just…

I mean, when you raise prices like
this, people that are, maybe they're

on infusion therapy that costs thirty
thousand dollars a year because they

have cystic fibrosis or whatever, they
have, they have an ongoing chronic

disease that they know they have
significant expenses, they stay in.

No matter what the price
is, they're staying in.

People that are healthy will not stay in.

And so then, you know, not
surprisingly, next year, they're

gonna have to raise premiums again.

And so you get in this like bad
spiral that I think, unfortunately,

I think we're already in.

Marcus: And then, uh, cybersecurity.

Gosh, so, so many challenges
in our industry, man.

Uh, this is Healthcare Dive.

Health experts say that healthcare
faces cybersecurity crisis.

These are patient safety issues.

Cybersecurity attacks
are, uh, are on the rise.

Um, they've brought hospitals,
clinics, and other providers

to the, to the breaking point.

Um, the healthcare sector consistently
ranks as one of the top hackers' targets.

Um, and, you know, DEF CON has
been happening, which is like the

big, you know, white hat hacker
conference, and they've been sort of

chronicling and talking about this.

It's a problem that's worsening.

These are not organizations that are,
quite frankly, that strong when it

comes to infosec and, and IT in general.

Um, they are losing the kind of
margin that you would need to,

to fend off these challenges.

And it, and AI is getting better
and better and better and more

widely available to, um, empower
the, uh, to empower the criminals

at executing these types of attacks.

And so we're not hearing a whole lot
about it because, as we've talked about

in the past, when some of the much, much
bigger hacks happened, um, a lot of these

folks just quietly pay and, and- Right

and deal with the fallout afterwards.

Um, but I, I th- I think If we look
at all of the trends as they're

playing out, uh, this is going to
be yet another significant area of

vulnerability for health systems.

Vic: Yeah.

No, no question.

I mean, we just covered that they're gonna
be under a lot of stress financially.

It's a lot-- It costs a lot of
money to, to set up all these

defensive cybersecurity measures.

And one, they don't have the talent to do
it, and two, even if they could find the

talent, they don't have enough capital,
enough, enough operating income to do it.

And then three, they-- typically,
most health systems use a pretty large

number of outside IT vendors that
are-- they're buying solutions from.

You have then that sort of derivative
vulnerability where one of them,

one of them gets hacked, and they
r- they ride in through that.

So it's, it's a perfect
storm of these things.

I think it's gonna be An
issue for several years.

Marcus: All right.

Uh, HHS moves to decertify organ procedure
group over patient safety failures.

So this is in the health system space.

The decertification of Network for Hope
could disrupt organ donation in Kentucky,

Indiana, Ohio, and West Virginia.

Um, I was actually with the founder of
transplant.org, um, a friend of mine.

His name is Tristan Mace.

He's a venture capitalist and, um,
transplant recipient, um, and survivor

of many in-incredible health challenges.

Just a, a great guy, and he's been
educating me about, um, all the challenges

and inefficiencies and, and opportunities
in the organ donation network world.

I certainly don't know much about this
case, and so, uh, I th-- I think we

have to, at face value, trust that
HHS has found some, some really, uh,

troubling matters here in, in terms
of, like, the way that they've been

dealing with, um, with these organs and,
you know, sort of persistent patient

safety failures is, is a big deal.

We need these networks to operate
at a very, very high level.

These, these organs are precious.

Um, but I can just tell you based on
my conversations with, with Tristan

that th-this is a really hard space.

It's not well-funded, um, and, and I think
that there are lots of vulnerabilities

in it, and decertification is, is
probably the right move, but I worry

about, you know, what that means in
terms of the accessibility of organs in

these, in these different states, right?

It's, it's like it's already
really, really difficult.

Um, Network for Hope say they
strongly disagree with the

decision, and they plan to appeal.

Um, but yeah, just a, just a
very unfortunate situation.

I, I hope this gets remedied quickly
for all those folks who are waiting

for an organ that are in those states
where Network for Hope was their

primary network they were depending on.

Uh, but yeah, just uncanny 'cause I
was just talking to Tristan yesterday

about all this stuff and, and
really pretty, pretty scary stuff.

Vic: Yeah, I mean, I agree that we, we
want to use the organs that are available

to help people, you know, get new organs.

Um, no question about that.

And in this article, HHS concluded
that at least twenty-eight times,

twenty-eight different people Were
not totally deceased when they

started the process of harvesting
the organs, and they can't have that.

I mean, that-- You, you can't say… I
mean, the example that they give is a, a

man overdosed and then was questionable,
but seemed to be regaining consciousness

and coming back, and yet they still, you
know, took the organs for other people.

That's insane.

We can't cross that line.

I mean, I think we have, we have
to- If, if-- I mean, I, I think

we have to trust that HHS is
investigating this in an honest way.

And if, if they found that twenty-eight
times, we, we can't do that.

That's not appropriate.

I mean, we need a lot more organ
donors, but it has to be, you

know, either voluntary or after,
after the person has actually died.

Sort of obviously, but, but, um,
it's shocking that twenty-eight

times is, you know, way too many.

It's not like one questionable case.

That's a whole series.

So- Agreed … I agree.

You know, decertifying and then
let's figure out maybe it's this

group reconstituted or a new group.

We should have an organ donor
affiliate program in these states,

but they have to be done properly.

It's just not appropriate.

Marcus: Yeah.

Ag-agree.

All right, moving into pharma.

PTC Therapeutics, uh, has a two
hundred and eleven million dollar

bid that wins, uh, the bankruptcy
auction of Sangamo gene therapy.

They, uh, Sangamo Therapeutics filed
for Chapter Eleven in June after their

search for strategic alternatives failed
to find a path forward for the company.

Um, th-- you know, unfortunately,
they were not one of the ones that's

been picked this year in all the
M&A frenzy that's been happening.

Uh, but the assets were, in fact, won
in bankruptcy and ex-extend the life of

those assets, which is, which is great.

That's a great, that's a great
outcome, you know, for advancing

the, the science and the technology.

Vic: Yeah, and this is how the
bankruptcy process should work, right?

It didn't-- You know, the science
didn't come out quickly enough, or

the trials weren't right, or they
spent too much money to get there.

They go into bankruptcy protection, but
then out of that, they're able to sell the

asset to another party with, with stronger
balance sheet that can carry it forward,

which is what is-- works out best.

Not necessarily for the invest-- equity
investors, but, but at least we could take

the IP and utilize it in another place.

Marcus: Yeah.

Certainly sorry for the
equity investors on that one.

Yeah.

Um, Braveheart Bio's three hundred and
eighty-two million dollar IPO for heart

drug leads to a big week for biotech IPOs.

Uh, more good news for the biotech space.

Um, this lead drug could be
competitive with, uh, blockbuster,

uh, Bristol Myers Squibb medicine
for a type of cardiomyopathy.

Um, another IPO.

Uh, looks like it, it landed well.

It was priced well.

Uh, I mean, Vic Are we in the wrong space?

Vic: Yes.

I mean, right now, biotech
is the, is the darling.

And, um- … there's a-- I mean, I
care a lot about heart disease, and

every week there's three or four new
technologies coming out, and then

this one just went public this week.

It's, it's an exciting time.

Lots of, lots of science going on.

I mean, it's one of the places where
you can deploy risk capital and there's

a, there's a pathway to s- to see
an exit, which I think is-- it sort

of all feeds together, as you know.

I mean, if you see an IPO, then, uh,
VCs are gonna now be i- you know,

more excited about investing in
those up-and-coming biotech deals.

Marcus: Indeed.

Indeed.

Uh, all right, health and us, and also our
transition into our technology segment.

"The data maxers feeding their
every health move to AI."

This is Wall Street Journal.

"From marathon training to tracking
office angst, health obsessives are

linking their data to chatbots to build
hyper-personalized coaches." I've been

on an interesting journey personally.

I f- I feel that when I was, uh, competing
athletically, uh, in jujitsu I was

definitely much more focused on tracking
and capturing data, and I thought it was

gonna be an obvious next move for me as
AI, uh, rolled out to say, "Of course,

why don't I pass all these signals
into these artificial intelligence, uh,

uh, applications and see how it can,
y-you know, give me more insights?"

I did do it a little bit, um, but I have
had real concerns over data sovereignty.

I've found myself actually, I think
for a combination of reasons, you know,

one being it's been, it's been more
of a busy work season than it has been

a competitive athletic season for me.

Um, but I think also because of
my, my growing concerns about

dai-data sovereignty in the AI
space, uh, I-I-I've been withholding

what I am sharing with, with AI.

You know, unless it is on a locally hosted
mo-model or, you know, maybe it's on a, on

a model on something like Venice, right?

Where, where I- Yeah … I have some
belief that, that the communication is

going to be, um, private and anonymous.

Uh, I-I've not been putting my health
data into Claude, Gemini, OpenAI.

What, what about you?

Are you, are you feeding
your data into those systems?

Vic: No, same.

I, I have not-- I mean, I, I do a ton
of AI work, and I'm also interested in

my own biometrics and tracking my stuff,
but I will not put it into any of the,

the, you know, frontier models or any
way that I don't have control over it.

I'm on Venice as well, and I've
had a couple questions I've

asked Venice, but I don't know.

I feel like your personal health
record, your genomics, all your

microbiome, all that information is
super valuable, and I plan to use

it, but I, I think I at least wanna
keep it where I own it, I control it.

I then-- maybe I'll have partner that I
let use it for, for a minute, but, uh,

just giving it to Anthropic in order to
get a quick answer that's not actionable,

like, you know, I'm not gonna really
do that much with this information.

There's no actionable step.

It just seems like feeding
them data for no real benefit.

And I think I've, I've learned from
social media, you know, y-if you're

not paying for it, I mean, maybe you
pay twenty dollars for, for, uh, uh,

OpenAI or Anthropic or something,
but if you're not paying a lot,

you end up-- you are the product.

They're gonna then monetize
you some other way.

Marcus: Yeah, and fun too, we've got
a company called ThriveAI Health,

which is a health coach application.

Uh, they are partnering directly
with healthcare organizations, right?

And, and I-- and that to
me feels More appropriate.

You know what I mean?

Yeah.

Like, it's, it's, it's taken with
the seriousness of, of what we're

dealing with, which is our data.

Now, don't get me wrong.

I want data sovereignty,
meaning I want to own my data.

But I would be far more comfortable
either putting it into a self-hosted

model to get feedback- Right … or,
you know, into some type of health

coaching application that was provided
by my health system, for example, right?

Right.

You know what I mean?

Or, or my PCP.

Like, that to me feels more in
the scope of how I'm trying to

manage this, you know, my data.

A- and, and, you know, look, I think, I
think things like, like 23andMe, uh, have,

have scarred me on this front, right?

Yeah.

You know, just sort of realizing that
just as our Social Security numbers

and all these other sorts of things
are just out there at this point,

you know, I'm not that excited about
my genome just being out there.

You know what I mean?

Right.

Vic: Yeah, and 23andMe is a good example
that they were pretty good actors until

they ran out of money, and then they
had to s-- you know, they had to sell.

And so you're totally at the
whim of, you know, they're out of

business, they're going through
bankruptcy, and who's gonna buy it?

And so even though at the time it made
sense, you never know what's gonna happen

with these companies, you know, over-

Marcus: Yeah … over the long run.

Agree.

Agree.

Uh, okay.

Moving into Web3.

I thought this was a very,
very, very big story.

Tether, uh, which is the, the, the,
the global juggernaut of stable coins,

um, USDT far predating, uh, Circle's
USDC, but also kind of not the, not

the, the most US-friendly, uh, stable
coin and, and also a, a stable coin

that people for a long time have
questioned the balance sheet of, right?

They've, they've questioned
whether or not they really have the

assets underneath, uh, the stable
coins i-in a one-for-one manner.

There's been a lot of talk about
it not all being, um, treasuries,

that they've got some gold in there,
they've got some Bitcoin in there, and

it's a pretty, you know, um, m-maybe
relatively esoteric balance sheet

compared to, say, a Circle, right?

Where, where it truly is sort
of treasuries underlying, uh,

their, their stable coins.

So with all that said, I think it's a
really big deal that KPMG issued a clean

opinion in Tether's first full audit, uh,
of their, uh, of their financials, right?

I mean, that's-- this is a real huge step
in them moving in the direction of trust,

um, here in the US specifically, right?

You know, trust- Yeah … that,
that those USDT coins that you

have, uh, are-- the bottom's not
gonna drop out of them, right?

They, they actually are, um, a
tokenized representation of a

dollar that, that you can count on
to, to be a dollar in the future.

So huge congratulations to them.

I'd had no idea this was even
underway, um, but this is a big deal.

Vic: And I think it's really, um,
because we are now opening up to, to

crypto Web3 assets, specifically stable
coins, Tether wants to get access to

the US market, and so they need to go
through this kind of audit with KPMG.

That was my only question,
like, who did the audit?

And KPMG is a, you know,
one of the Big Four.

They, they're clearly-
Big Four … a g-- Yeah.

And, you know, in year one of
Tether, which was probably, gosh,

ten years ago, eight years ago,
they, they grew incredibly quickly.

Um, I know 'cause I had personally
funded a stable coin here in Nashville

that was maybe more elegant, but much
fast-- much slower to get to market.

Um, so there were a lot of questions
about how quickly they were growing

and could they possibly have, you know,
everything properly, uh, reserved.

But at this point, they,
they're a huge operation.

I mean, multi-billion dollars in
annual income just from the f- from

the flow off the, off the earnings
And so I think they have plenty of,

uh, capacity now to be fully reserved.

Um, but I agree with you completely.

Having KPMG give them a clean audit is, is
very different than me saying, "They got a

bunch of money, it's probably okay." It's,
you know, much higher, higher validation.

So great- Absolutely.

I think it's healthy that
Circle has a competitor.

I mean, we need, we need
competition just to have it, it

be-- it'll keep them both honest.

Marcus: Well, look, Tether's a
hell of a competitor, I mean-

Vic: Yeah.

Marcus: Uh, you know, that thing
is l- that thing is as close to a

money printing machine as, as one
could possibly hope to operate.

Yeah.

Um, so that's a really,
really big competitor.

And if they, you know, if this audit is
any indication of what they are looking

to do moving into the United States
seriously with, with, with their, their

market share globally, with their balance
sheet, uh, they're b- they're gonna be a

very formidable competitor for, for sure.

So.

Vic: Yeah.

And I mean, Circle has
plenty of resources too.

So I think they-- it'll be good for me
as a user- Mm-hmm … to have them both

in the US, both, you know, competing
for, for my stable coin, you know,

my hundred dollars in stable coin.

Marcus: Yep.

Uh, okay.

And also here from CoinDesk, uh, the
SEC has canceled their long-awaited

proposal of Reg Crypto, um, postponing the
meeting, importantly, without a new date.

This is the, the continuing
saga around the Clarity Act.

The SEC and the CFTC have been messaging
that, A, we need the Clarity Act,

but, B, if you do not, uh, bring the
Clarity Act into reality, um, then

we're gonna have to create these
regulations at the, uh, agency level.

And not necessarily what we, what we
want to do, um, but clearly they're

incentivized to do this because this is
what the, the crypto lobby wants from them

and expects from the Republican Party.

They expect them to deliver, uh,
not just genius, but also clarity.

Um, I think that there's been a
real hangup around the, uh, the

allowance of yield on stable coins
that is embedded inside of Clarity.

Mm-hmm.

When most of us think about Clarity,
we're thinking about the market

structure component of it, right?

Which is the ability to offer
securities via tokens, and then take

those tokens across the blockchain and
move, um, move with them from there.

And but that's not the only
thing that's inside of Clarity.

Mm-hmm.

There are other things in there.

That's right.

And I, I think the bank lobby
is really driving, um, a lot of

opposition around that, that,
uh, that stable coin yield issue.

So who knows what's going on here.

You know, maybe them pulling back is
a signal that somebody's whipping up

enough votes to, to move Clarity forward.

You know, maybe that's
what this actually means.

Uh, but we continue to have, uh, a lack
of clarity around whether or not this,

this bill is gonna make it through.

Vic: I agree with you.

There's something going
on behind the scenes.

I don't really know what it is And I'm
hopeful that we get some clarity around

the Clarity Act or with SEC or CFCC, but
it-- I'm not confident that they're gonna

get something passed before the midterms.

And so we're looking at next year.

Now, maybe I could be wrong about that,
but it just seems like a lot to do in…

You know, we're already in
sort of the campaign time where

they're not that focused on it.

So I don't know.

Disappointing, but we'll
see what, what comes of it.

Marcus: We've got four
AI stories to wrap up.

The first one comes from The Verge.

It's, it's chronicling the shakeup
that has happened at DeepMind.

Um, and the headline is, "Does
Google even want to win at AI?"

And I actually think that this
is the right question, right?

So we have had Jeff Dean, who
was DeepMind's chief scientist,

leave to form his own startup.

The CEO of DeepMind, Demis Hassabis,
is stepping aside to focus on,

quote-unquote, "longer term research."
Um, and several other key, uh, key

staffers at Google that were heading
up their AI efforts have left.

You know, the, the truth is,
it really doesn't matter what

other things you have going on.

In the AI world, talent is king, and when
you have an exodus of, of people at this

caliber leaving, um, it says something.

And certainly when we talk about Google,
we're not talking about an organization

that does not have the infrastructure,
the balance sheet, the, the

capabilities, the data to, to win at AI.

Um, what people are rumbling about
is whether or not Google has the

culture or the will to win at AI.

Um- It sounds like Sergey Brin has
now stepped in to, to deal with

this, this vacuum, uh, of power.

And, you know, whenever Larry
or Sergey have stepped in in the

past, they've actually been quite
effective at righting the ship.

You know, Google is fortunate to
have these founders that can kind

of swoop in, uh- Right … whenever
there is a, a bunch of melee and, and

put things back into founder mode.

And so I certainly am
not writing them off.

I also think, and I continue to just
sort of point to, okay, we're talking

about these model companies, and their
financials look all crazy, and every day

I'm out in the streets here in Nashville,
and there's Waymos on the street, right?

So to me- Right … it's like the,
the-- I, I don't really know what we,

we, what we mean when we say, "Does
Google even wanna wi-win at AI?" when

they are the company with the most
autonomous vehicles on the street.

You know what I mean, like- Yeah … that
to me just seems sort of insane

that we continue to look at these
benchmarks around these models alone.

But Vic, y-you know, you're,
you're studying all this.

What's your view on it?

Vic: Yeah.

I think we're pretty aligned.

I mean, I, I think Google
has a lot of assets in AI and

outside of AI, as, as you said.

At least my opinion, and so I'm biased
'cause it's my opinion that I'm now

projecting on Google leadership, but
I'm not sure you have to have the, the

leading frontier model to win at AI.

I mean, they are winning in the cloud.

They're winning in the, you know, sort
of the usage with all their applications.

They have their own chips, and they
have a, a very good AI, and then they

also allow you to use whatever model
you wanna bring on their systems.

I don't know.

I think that they might be sort of
playing a different strategy than,

certainly than OpenAI and, and Anthropic,
who don't have those other assets.

I'm not sure that equates to
they don't wanna win at AI.

They're just, they're just playing a
slightly different strategy out because

they have a different set of assets.

Marcus: Yeah.

No, look, I, I, I agree.

Uh, I, I think that I, I think that
Google is playing an, an, a much more

integrated game, and they have many
more places where they can apply AI.

They're not just this
model harness company.

Um, I, I, I wish they were
better at it, you know.

I certainly wish Gemini was-- had
a better harness because we already

pay for Google Workspace, right?

It's like-

Vic: Yeah


Marcus: if, if, if Google took this
whole harness aspect of, of AI more

seriously, we wouldn't even need Claude.

Vic: That's right.

Marcus: You know?

'Cause our business-
Yeah … lives on Google Workspace.

Um- Right … but they don't.

They don't take it that seriously.

I, I-- So, um, I, I think it's an
interesting question, this whole,

like, do they even want to win at AI?

It's like when we say, "Do they want to
win at AI?" We're really talking about the

harness business model, I think, you know?

Um- Yeah,

Vic: I mean, I think that- We're-
Yeah, I think that's right.

The, the-- Sorry, I didn't mean to cut
you off, but the, uh- No … the recursive

self-improvement is the one sort of place
where I think you can make an argument

that if Anthropic or OpenAI start to get,
you know, even a s-- you know, somewhat of

a lead, and then they get-- they use that
best model to teach itself more quickly,

where they could really get, get ahead
and be pretty dominant, where it would be

really damaging to Google's market cap.

At the end of the day, that's
what they're focused on.

Um, it seems to me like it's evolving
now where there's, there's at least

two leaders, and then there's a group
of maybe ten other, if you go to China

and Groq and Google and, you know, um,
there's a couple choices in Europe that,

that are, you know, are hanging around
that we probably won't have a runaway,

you know, single winner to lead all.

Um, but that's their, that's
their vulnerability, I think.

If that happens, it could be damaging.

Marcus: Yep.

Agree.

All right.

OpenAI sheds senior
execs in pre-IPO refresh.

So there's lots of things moving around
in, in teams- Yeah … right now.

Um, OpenAI's Chief Revenue Officer,
Denise Dresser, is leaving the company

less than a year into the role.

Um, her exit comes the same time as,
as, uh, Bright-- uh, Brad Lightcap,

who was the former CEO, um, ha-has
announced, uh, his departure.

Uh, just lots of, lots of moves
happening around these labs, and I

think the competitiveness, the amount
of capital that's being thrown in and

sort of the, the winner-take-all sense
around this race, can you imagine how,

how hard it must be to make forward
progress and retain talent right now?

Vic: Yeah.

I mean, it, it is the ultimate
sort of competitive game, right?

You have two really well-capitalized,
incredibly talented teams in, in

OpenAI and Anthropic, both pushing to
get to an IPO in the next six months.

I think Anthropic's probably
marginally ahead in that process.

And unlike Google, they don't
really have alternatives, right?

This, this, th-this is the one
thing that they're gonna do.

Yeah, she, she was-- I mean,
Denise, who, who was the head of

sales, she was there 10 months.

I mean, that's the
competitive landscape, right?

She gets there, it takes her two months
to figure out, you know, what- … what

the heck she's selling, and then
eight months later, she's gone.

I think that's just the
environment that we're in.

Marcus: It's insane.

It's insane.

And, and I mean, look, that, that
churn is also inefficiency, right?

So you think about, like, all the
capital that's being absorbed in

this space, some percentage of that
capital is going to the inefficiency

of being able to retain a team.

Vic: Yeah.

Marcus: Right?

The constant spinning up and shutting
down of these different groups

of people working on, you know,
different aspects of these businesses.

It's, it's insane.

Vic: Yeah.

Marcus: It's insane.

Uh, and you know what?

I bet you that that is internally
really driving these labs to wanna get

to recursive self-improvement, right?

Because Yeah … I bet they're just
like, "Screw these people." Yeah, yeah.

"We, we, we, we gotta get to where
the AI is smarter than the people-"

Yeah "… so, so we can stop dealing
with these people," you know?

Vic: Right.

Right.

Marcus: And their need
for prestige and money.

Uh, Decrypt, Google and OpenAI
debut super fast AI models.

Gemini 3.7 Flash is out, but GPT
5.6 Sol Ultra Fast is invite only.

Um, I was listening to the Limitless
podcast, and I really liked one

thing they said, Vic, and I wanted to
bring it up on this show, which is,

we talk about the frontier, right?

Um, but obviously, you know, DeepSeek
has had a, a monster couple of weeks

here, and it's, it's like we're
at the point now where when we say

frontier, what do we really mean?

Because there are n-now
different frontiers, right?

Yeah.

There's, there's, there's the
frontier of, we'll just call

it general intelligence, right?

And nobody's quite there yet, but that
is, that is pretty much what Anthropic

and OpenAI are duking it out for, right?

They're, they're, they're-

Vic: Forget cost or speed.

Just what, purely intelligence,
who's the best, who has the

best, most smartest model?

Marcus: A hundred percent.

A hundred percent.

Right?

But there is also, like, a
frontier of size, you know?

It's like Meta and, and, and the
Muse Glimmer, they're, they're not

going for the most intelligent model.

They're going for the model
you can run on your laptop.

Right.

Right?

Um, there's the frontier of speed, and
it does seem that in fact, like, Google

is very interested in speed, right?

Which, which makes sense because,
you know, Google's always been

about trying to push more and more
things, um, to the browser, right?

The- Yeah … they've, they've
always optimized for speed, and

so it makes sense, you know?

There's, there's gonna be the,
you know, the, the, um, the, the

frontier of, uh, specificity, right?

The- there's gonna be small models
that are not generally very intelligent

but are trained or maybe even tuned
to focus on specific domains and maybe

even down to specific tasks, right?

So, so the frontier, when we say
this, the, the frontier, it's

like- That's an evolving concept.

Vic: Yeah.

And it's, and it's jagged, right?

Like there's some places, say
software, where many models

are far better than a human.

Then there's other places, intuition
or like understanding what someone

means, uh, when they, when they talk
about an experience they had that, uh,

is harder for the LLMs to understand.

Then you get to the whole world model
where they're trying to understand like

the spatial three-dimensional things.

So it's not-- I don't think
it's a one-size-fits-all.

There's, like you're saying,
there's different use cases and

different models have different
flavors or different specialties.

Uh, the one other axis that I
might throw out there is, is

like the value axis where the-

Marcus: Mm-hmm

Vic: it's like a combination of quality
but also cost per task, which isn't

exactly the same as cost per token.

Um, you know, some of the Chinese
models are very inexpensive, but

they use two times, three times as
many tokens to do the same thing.

So you have- That's right … to
sort of balance all of these things.

Uh, we already have moved to,
you know, kind of routing the

different jobs to different models,
and I think that's, that's really

gonna be the late 'twenty-six,
'twenty-seven thing in my, in my mind.

You have a harness, and then you're
routing to a whole bunch of different

models to get the combination
that you need for this, for this

situation and this one, this one
task I'm trying to do with today.

But tomorrow, I might
need a different model.

The other thing I, I think is just
interesting is that the invite-only

thing, I think we're gonna see more and
more of that for sort of two reasons.

One, I think that it's becoming
harder from a regulatory point

of view- Mm-hmm … to release
models, um, that are unconstrained.

And then secondly, I think that
it's just true that there's a

commoditization in the price per token.

And so I think they're gonna start
gating some of the best, the best

value or the most extreme intelligence
models either for their own use, and

let's spin up a hedge fund 'cause we
have the best model in the world and,

and we're not gonna give it to anyone.

We're just gonna use it internally.

Or an invite only where they are, you
know, giving access to a group that either

maybe they're paying a premium or they're,
they're giving them some data in a trade.

Um, I think OpenAI and Anthropic
particularly They have to

consider all these options.

There's no, there's no rule that they,
they must release every model they come

up with, and there might be other ways to
make revenue that are more strategically

valuable than selling tokens.

Marcus: Yes.

Yes.

Very much agree.

Uh, yeah, great points on both cost,
uh, cost/value, uh, and also the,

the need to, um, gate who can, who
can access these models, for sure for

distillation reasons, but I think the
regulatory argument is also important.

So, um, okay, final story.

GrokBot was announced on August 11th, and,
uh, you know, this is one of those things

where we started with Open Claw I feel
like the next big thing was, um, Hermes.

Mm-hmm.

Then Perpe-Perplexity Computer.

Vic: Yeah.

Marcus: Then most recently Buzz,
and now we have GrokBot, um, the

team of always-on agents, you know.

Uh, it's getting to the point where
y-you just really can't count any of

these organizations out, especially
if they've got meaningful, meaningful

resources and they've got some type of
distribution and data capability and,

and that's-- for me, that's specifically
talking about like Meta and, and, uh, XAI.

You know what I mean?

Yeah.

It's like these guys have-- Because of
their distribution, they've got so many

different ways they can go about this.

They can repackage AI and put
another meaningful shot on goal.

Look, they're not gonna beat OpenAI
or Anthropic at the, quote-unquote,

"frontier of pure intelligence."

We know that now.

But the application of intelligence in
ways that is actually meaningful and

helpful for their, their audiences, I
think they actually can be, you know,

quite effective in, in, in those areas.

Have you used GrokBot yet?

Vic: No, but I a-agree completely.

So, you know, you're right.

It came out on Tuesday.

I'm planning to download
it this weekend and try it.

What, what I'm excited about is
it, it is very similar to Manus.

So Ma-- I don't know, I mean, if you
recall it, Manus was one of the first

agents that were re-- that was really
easy, usable, and it was in the cloud.

The, the big knock on it was it
was, it was a Chinese company.

Maybe I guess it is a Chinese company.

It got bought by Meta, but
then I'm not sure if they ever

were able to actually get it.

Um, and this, I mean, just from the demos
I've seen, it's very similar to that.

It's meant for a non-developer, right?

So you basically-- It's in the cloud.

There's no, there's no
installation on your hard drive.

I mean, you download the,
the app, but that's it.

And then it, it does everything
kind of behind the scenes, which

is what I liked about Manus.

You, you could just tell it, "I
need this task done, and I don't

really care what model you use or
how you go about it. Just, you know,

let me know when you're finished."

Marcus: Right.

Vic: They have taken that same kind of
experience for the general population.

I mean, you and I are playing with
all these tools, but most people,

they wanna be able to pull out their
phone and, and ask for something, you

know, like, "I'm, I'm going to lunch.

Can you do a competitive
analysis on these four companies?

'Cause I'm invested in one of them,
but I wanna compare these other

three." And then I go to lunch,
and I come back and it's done.

And with OpenAI and Claude and Gemini,
you still have to sort of be there

sitting there with it and saying yes
to permissions and, "Okay, yes, let's

take that next step." And I haven't
tried GroqBot, but I th- I think their

use case is much more like we will take
all this complexity, all the technology

stuff away, and if you don't like the
results, then tell us what part you

don't like, and then we'll learn- Mm-hmm

from that and iterate.

Like, you don't need to know… I
don't think you can pick the model.

It just, it just, it, it's using Groq.

Like, they're using whatever model they
want, and if you're not happy with the

results, then you shouldn't use it.

But I think a lot of people will like
that kind of easy button of just, you

know, just give me the answer, just
give me the output, and I need that.

Marcus: I, I, I think, I think that's
right, and I think that that is a, that's

a clear differentiation between all of
the previous models that I talked about,

the, you know, whether it's OpenClau or
Hermes, Perplexity, Computer or, or Buzz.

In those, you are selecting, you know,
you're selecting the, the, the model.

You're sel- you're even
selecting the endpoint.

Um, in this one, it's just all packaged
up, and I think it's interesting-

Yeah … that, uh, that as Manus has,
uh, been removed from, from Meta's,

uh, platform and y- you know, therefore
now back in, in Chinese hands, and I

think going to be a much harder company
to commercialize in the United States.

Yeah.

It's interesting that here
comes GroqBot to replace it.

Vic: Yeah.

Yeah, and I think it's a signal that we
might be moving past the, like, early

adopter tech tinkerer phase of AI and
getting to where there's, there's a lot

of people that they wanna use AI, but they
don't wanna learn all this other stuff.

They, they don't wanna open up Terminal.

And the stuff that I love about Claude
Co-Code, it makes my wife's head hurt.

Like, she, she doesn't wanna do that.

She, she just wants to have, like, an
easy thing that answers stuff for her,

and that's a much bigger market if it--
But, you know, if GroqBot can do it at

high quality, that's the big question.

Marcus: Yes.

It, it is a much bigger market.

It is also, um-

So thi- so this is a really
interesting thing, Vic, right?

So in other spaces where we move from the
geeky, I wanna tinker, I wanna control

everything to just give me the… So
let's, let's just take the difference.

It's, it's not a perfect analogy,
but let's just take the difference

between like Linux and macOS, right?

Yeah.

They both run on this sort of
Unix-based operating system, but

like one is just an absolutely
terrible user experience, right?

And the other one is, is
pretty powerful, right?

The other one is pretty powerful.

Um, o-one of the things I've been
really disappointed in is the lack of

alternative harnesses out there, right?

So like I, I, I keep wanting to move
away from Claude or OpenAI to one of

these other models, but these, the-these
harnesses are just not up to snuff, right?

They're, they're simply not up to snuff.

And- Yeah, it's

Vic: like, it's like
spinning up a Linux server.

I mean, it's

Marcus: hard.

Yes.

It's… Yes.

It's really, really difficult, right?

Um, and I, I think the question I have is

What is the, what's the trade-off gonna,
gonna ultimately be when we're not dealing

with, um, ease of use only in terms of
user experience, user interface, right?

But we're talking about not only
even quality of performance of these

models, but also like what these models
are trained to return to you, right?

Like, you know, so what, what gives
me pause about this is that it's Grok.

You know what I mean?

Right.

And, and, and for better or worse, I
know what Elon Musk's personality is.

I-- You know what I mean?

And so, like, do I want

Vic: to- And I mean, so far he has
imbued that personality into Grok.

I mean, it, it has a personality that…
I mean, it's like, it's like being on

X or like hanging out with a, an Elon,
I don't know, clone or a different kind

of entity that's sort of like Elon.

Marcus: Yeah.

So, so, so it's like, do I want to
trade, uh, complexity for simplicity

and in that trade take on, you
know, always in the back of my head

knowing that this thing is spitting
out what Elon wants it to spit out?

Mm.

Vic: Right.

Yeah.

Marcus: So that's, that's the, that's,

Vic: that's the challenge.

That's, that's a concern.

That's certainly a concern.

I think that, um, this, like,
crossover to ease of use mass market

does not work without a killer app.

Like, so, so I just don't see… Like,
until the-- Like, people don't have enough

things of value to give Grok Bot now
'cause, 'cause we haven't figured out,

okay, what, what is the thing that will
save me five thousand dollars a year or

five hours a week that Grok Bot can do?

Marcus: Yep.

Vic: And that might get invented over the
next couple months, in which case Grok Bot

will take off if there is a killer app.

But right now it's just kinda you can
do all the same stuff could do with

other things, but it's just easier.

And I think that, uh, the mass market
needs to have like an easy button.

Like, and I don't know what that is,
but, you know, the iPhone… I mean,

macOS was always good, but then really
the iPhone was the killer thing that,

that just accelerated everything.

Marcus: Yeah.

Yeah.

I agree.

I agree.

And, and, and look, of the
companies out there, if we're

talking on the consumer side, uh-

X is a decent product company.

Yeah.

You know, you, you, you know, you, you,
you may not like the political , you

know, film that, that the product
flows through, but, uh, it- they're,

they're a good product company.

I mean, they, you know, they just
rolled out X Money and I'm, I, I

signed up just to, like, experience
it, just to see what it was like.

I don't know if you know
the- I don't know what

Vic: that is.

Okay.

Marcus: Uh, well, do, do, do you
have, um, do you have X Premium?

Vic: Yeah.

Yeah.

Marcus: Okay.

So you, you sh- you should see whether
or not you got an invite to, to X Money.

Okay.

I, I did.

I signed up.

They give you $25.

Hmm.

Yeah.

I mean, I'd sign up just for that.

I mean- For, for, for, for…

Yeah, exactly.

For the, for the early folks.

Yeah, right.

So, so you should, you
should check your app.

You should check your X app- Yeah … and
see if you've got early access to it.

Um- Yeah … but anyway, look, our job is
to just continue to track all the things.

So- Yes, right … let me know next
week when we record, uh, uh, what, um,

what's your thoughts are on GrokSpot.

Vic: Yeah,

Marcus: I'm gonna test
it out this weekend.

Yeah.

Oh, hold on.

Next week.

Uh- Oh, uh,

Vic: next week you're out.

Marcus: Yeah.

Do you have, do you have a guest

Vic: host?

Yeah.

Yeah, I have a guest.

Paul Kappelman, who's done it-
Hey … previously is gonna

do it, so that'll be great.

Yeah.

Marcus: Awesome.

Vic: Awesome.

We'll hear what's going on on the
private equity side of things, and just

catch up with Paul, so it'll be good.

Marcus: Love it.

Love it.

It's been a while since we've had Paul
on, so I'm looking- Yeah … forward

to listening to that episode.

All right, man.

Yeah.

Look, have a great weekend.

Vic: Okay, bye.