Subscribe
Share
Share
Embed
Artificial intelligence is shaking the foundation of the business world. Are you ready for what's next?
Welcome to AI: The Podcast, your weekly guide to the most important AI news and how it directly impacts your business, career, and industry.
Hosted by David Maples —an AI expert, intellectual property attorney, and CEO—and Virginia Huling—artist and creative director—this podcast bridges the gap between hard business strategy and the creative human element. Every week, David and Virginia cut through the noise to break down the biggest stories in Artificial Intelligence, offering unique insights from the boardroom, courtroom, and design studio.
Whether you're an entrepreneur, executive, creator, or tech enthusiast, you'll discover how to leverage AI tools, navigate complex copyright laws, and future-proof your career.
What you’ll get every week:
• The latest AI business news and tech updates
• Expert legal insights on AI and Intellectual Property (IP)
• The impact of Generative AI on artists and the creative industry
• Actionable strategies to implement AI in your business
• And more!
Hit subscribe to stay ahead of the curve!
Virginia Huling:
Welcome back to AI the Podcast, where we cut through the hype and have honest conversations about how artificial intelligence is actually impacting businesses today. I'm Virginia Huling.
David Maples:
And I'm David Maples.
Virginia Huling:
It is the week of May 4th. Before we get into today's news, David, I want to clue the audience into a conversation we had the other day about the general landscape of things. We wanted to address the elephant in the room: the growing backlash against AI.
For our audience, we want to make one thing crystal clear right out of the gate. This show is stubbornly human-centric. We care about the people who manage businesses, run businesses, and work in businesses, not just the algorithms.
David Maples:
It's interesting you say that because just last week we saw Jeff Dean, one of the chief architects at Google DeepMind, get booed off stage at UC Berkeley for Gemini’s participation in what’s going on in Gaza and the technology being used there. We’re starting to see these ripples surface in society.
I’ve been thinking about the show a lot over the past week. As a systems architect and an IP attorney, my job isn’t just to be a hype pitchman for the next new AI technology. I’m very pro-tech. I think there are ways to use it in your business. But ultimately, my job is to help you, through this podcast, build structures that protect your ideas, your data, and your people, because at the end of the day, it is about the people. Everything else is irrelevant.
Right now, in May of 2026, the landscape is shifting, and not just a little. It’s shifting aggressively and fast. Big tech and government regulators are rewriting the rules in real time to serve themselves. If you aren’t paying attention, human-run businesses are going to get squeezed and run over.
Virginia Huling:
Exactly. Today, we’re going to look at this massive geopolitical chess game happening right over our heads. It’s wild. It’s like history in real time. What will things look like 100 years from now when people look back on this time period?
We’re going to cover why the price of enterprise AI is about to skyrocket even while the models themselves are being commoditized by open source. We’re going to look at how new U.S. security regulations are actually a trap called regulatory capture. We’re also going to explore the very real threat of foreign open source models acting as Trojan horses.
David Maples:
There’s a lot of noise. As someone who lives and breathes what’s happening in technology every week, the amount of information coming out right now is not just drinking from a fire hose, it’s dizzying.
Our goal on this podcast is to connect the dots back to you, to show you how you belong in this architecture. The technology is just the plumbing. Human beings are the asset. Let me repeat that: technology is just the plumbing. Human beings are the important critical point at the center of all this.
Our goal today is to give you strategy to navigate these global shifts without losing the real, authentic human connection that your clients actually pay you for.
So with that, let’s dive right in.
One of the big things that happened is that it came out a couple days ago that the Trump administration is considering mandatory pre-release vetting for new AI models before they are released. What’s really interesting is that this is all under the guise of Claude’s Mythos model. The framing is: it’s too dangerous, nobody can have it, and now the government is stepping in at the apparent behest of the people who might benefit from it the most: the biggest AI companies, the ones talking about IPOs this year that might value them at a trillion dollars while still being negative in revenue.
What’s going on here is that the companies building and releasing AI models, and trying to build infrastructure to profit from them, are now asking the government to step in and vet those models before release.
Virginia Huling:
You said something about a year ago. You looked at me and said, “There’s no more moat.” Once you figured out what could be done with the models that were being released at the time, you skipped ahead and saw that everything would get watered down once people had access to these models.
So now, are they trying to artificially create a moat so there’s less competition?
David Maples:
They’re not trying to create a moat. They’re trying to build a wall, and a wall you have to pay to get through. There’s a gate, and they will manage that gate and make sure you need several billion dollars to get through it.
Virginia Huling:
Boil that down for me. We’ve got enterprise-level businesses that are obviously going to be utilizing this. I’m sure Salesforce is going to rope this into their system somehow. They’re probably working on that right now. But how does this boil down to mid-level companies and small businesses? What does this mean for the rest of us?
David Maples:
Two things. As the big players go, enterprise companies follow. Historically, we saw this with the FCC, with radio, TV, and the railroads. Regulatory capture is a way for a company to artificially build walls around an industry when it doesn’t have another way to protect itself.
What’s happening here is that AI is becoming commoditized, and it is going to cost an incredible amount of money to run these next-generation models. These models are not ten times better. They’re 20% or 30% better. Maybe they’re better for certain activities, but how many PhD-level scholars do you need going through your email inbox? Do you really need that, or could you just use the plucky high school intern equivalent?
For business owners, here’s the key point: this is textbook regulatory capture. If the government says these are the only models the federal government is allowed to use, and if you work with the federal government, then you are only allowed to use certain models. That creates massive vendor lock-in.
Imagine in the railroad industry you could only carry your goods to market if you used one specific train run by OpenAI, and it cost $1,000 per ton. Meanwhile, another train costs $10 per ton, but you’re told you can’t use it because the destination is a government warehouse. That’s the play.
Virginia Huling:
To play devil’s advocate, Mythos was the catalyst for this, and maybe for good reason. What is being presented is that Mythos scared enough people that someone said this is why we need this kind of regulatory setup and review. A lot of common-sense people would look at that and say that makes sense.
So, David, you’re making the claim that the timing may be coincidental, but maybe it’s also cover?
David Maples:
The private conversations in Silicon Valley for the past six to eight months have been about needing regulatory capture. When DeepSeek came out in January 2025, everybody said it was going to eat their lunch. It was only three to six months behind them, and China is going to keep releasing open-source models because they don’t have the chips and infrastructure to run the very best inference models, but they’ve made a very decent Chinese knockoff. Qwen and Alibaba’s models are great. I’m currently running Qwen 3 and Gemma 4 locally on silicon that doesn’t cost me a quarter million dollars. I can do this on a $10,000 rig.
In politics, there’s a saying: never waste a crisis. Mythos created a crisis of opportunity, and the AI companies were more than happy to use it. That’s what I’m saying. I’m not trying to be a conspiracy theorist, but they have definitely taken advantage of the moment.
Think about how the script got flipped. The Trump administration originally said no regulation on AI companies at all. Then Dario had previously asked for regulation. At the time, I thought they might get left out in the cold. It appears they already got a black mark.
What I’d like to say to the audience is this: you cannot blindly pump your client data into systems that are going to be intertwined with the federal government. You’re going to have to build a local middleware vault. You’re going to have to scrub, anonymize, and redact proprietary information locally.
Virginia Huling:
Why is that?
David Maples:
Because if it’s going into government databases, you lose control. The larger point is that you need to make sure you’re protecting your intellectual property, because once it goes into somebody else’s database, getting it back is hard. It’s like trying to lock the barn door after the horses are already out.
I also think business owners need to think hard about making their systems agnostic.
Virginia Huling:
We’re going to talk about that because you also mentioned that the price on these systems is going to skyrocket. I’m going to use that to roll into our next segment.
What’s interesting is that everything we just talked about is followed by major tech giants such as Google, OpenAI, and Anthropic formally agreeing to share their models with the U.S. government for security reviews and to fall in line with this new regulatory framework.
This consolidation of the relationship between federal regulators and Big Tech sets a national standard for what constitutes a “safe” model. But like you were saying earlier, this really looks like textbook regulatory capture.
For those listening who may not be familiar with the term, regulatory capture is an economic and political concept where a government agency is created to protect the public interest but ends up being controlled or heavily influenced by the industry it is supposed to regulate. Instead of acting for the benefit of society, it acts in a way that benefits the major players in that industry.
It’s really interesting that now the biggest players are working directly with the government. This threads right into everything you were just saying, David.
David Maples:
There are a few things to be said here, and I want to call it the way it is. The rising talk about potential conflict with China is a problem, but it is a matter of public record that Chinese espionage has stolen trillions of dollars’ worth of capability and technology from American tech companies.
For years, Google would not allow engineers from mainland China into certain positions involving Google search algorithms and specific areas of internal access. They were heavily criticized for that. Within six to nine months of changing that posture, China had allegedly stolen exactly the things Google had been trying to protect and built clone copies.
This absolutely happens. I think the temperature needs to be brought down a little bit, but the reality is that this issue depends on your perspective.
I heard Jensen Huang, the CEO of Nvidia, on the Dwarkesh Podcast. Jensen is not used to getting pushback from someone technically sharp. The host challenged him on whether supplying advanced AI chips to China is like supplying uranium to a nuclear program. Jensen rejected the comparison, but in reality, if you believe AI represents an existential threat, then the analogy is not unreasonable.
Every one of these tech leaders is self-serving. Jensen would love the Chinese market to be open. He would love to sell chips there. The consequence is secondary to the business incentive. That’s his role as CEO, but it is potentially dangerous, and it requires thoughtful people in government making decisions about it.
The same thing is going to happen here. People are going to argue that Chinese open-source models are a risk. There are risks, but those risks can be mitigated. For most small and medium businesses, you are going to need a company like ours to help build the infrastructure to run that stuff on your own silicon. That is achievable, especially if you can’t afford the buildout for enterprise systems.
Speaking of China, I’d like to go to our fifth article. This is from The New York Times, and it’s about how AI is transforming China’s entertainment industry. Virginia, you predicted this two years ago.
Virginia Huling:
Micro dramas. Hear me out.
The story is actually really interesting. Like David said, it’s a warning for Hollywood. Humans create massive amounts of content every day for the entertainment economy. We’ve gone from books to plays to television to movies, and now to in-hand entertainment on phones.
Creative people are starting to mass produce what are called micro dramas. Think of them as soap operas delivered in 30- to 60-second segments, stacked together so the story unfolds in sequence. The goal is to capture your attention and trap you in a compelling little story, and much of this is being created with AI.
That’s not universally true, because people are using AI in different ways. The article covers different viewpoints, including a creator, a director, and a producer. You get to see how the creative side of AI is allowing people to tell stories in a way that is much more accessible and much faster to produce, while creating a viable industry around it.
It’s really interesting. On the whole, I know AI represents a massive creative threat to individuals and filmmakers, or makers in general. But this article also shows some of the opportunities it creates.
I say that not to ostracize my fellow creatives, but to remember that AI is a tool, and we are still in the early days of this technology. There is a lot of push and pull, and a lot of debate over whether we want it or not. Personally, I have a love-hate relationship with it every day.
But when I think about people who have a story they want to tell and haven’t had the ability to do so until now, I find it incredible that this technology is at their fingertips.
David Maples:
I am very torn on this, and I’m going to put on my human hat for a minute. I think this is absolutely a race to the bottom.
The article covers how, in China, entire film crews are being replaced by AI models. They’ve reduced the cost of VFX, scripting, production, actors, directors—everything—to about $30 a minute. The result is a lot of slop, and it is facing backlash.
Virginia Huling:
Due to volume alone, it’s going to create a ton of slop. Within all of that, though, there will be a few diamonds.
It’s the same reason people used to ask us, “Can you make my video go viral?” The answer was always that if we produce enough content, eventually something might go viral. There is no formula.
If people are just trying to produce a bunch of junk content, that’s obviously going to be rejected. But the potential is there for a different level of creation. Maybe there really is an opportunity for people who want to be good storytellers and just haven’t found the right medium yet.
David Maples:
No doubt. I think there are good uses for AI in efficiency. For example, we have producers for our podcast who do color grading. If AI can replace repetitive production tasks like that, great. I want their brains reserved for the more human parts of what they do.
Here’s the bigger point: when content and code generation cost zero, the output becomes worthless. Competing on AI efficiency alone is a losing game. Businesses should use AI for back-end efficiency and for things they couldn’t do before. But for client-facing products, I think you need to keep them deeply human and deeply personal.
One thing we’ve talked about recently is that professional services firms should remove cents from their pricing. Round numbers communicate premium, human-driven strategy. They imply that you’re paying for a nuanced strategic mind that an algorithm cannot replicate.
The problem with AI slop is that it removes the human from the process. When you eliminate human interactions and touchpoints in the race to produce the cheapest possible service, I think that is a losing strategy.
Virginia Huling:
Any time you try to mass produce a bunch of low-quality content, it’s going to be rejected. What I’m saying is that the potential is there for a new kind of creative process.
That said, there are definitely people creating junk just to get it out there and monetize it. I got caught by one the other day. It was this sad story about an older woman signing paperwork to surrender her dog at a shelter. I started reading it and thought it was heartbreaking. Then I realized it was fake. It was AI. And now I’m starting to notice that pattern more often.
So yes, people are definitely creating manipulative content just to attract attention and make money. But I don’t want to completely dismiss the possibility that some people will use this well.
David Maples:
There was an article a few months ago about ads being run completely by AI, which is where Meta is trying to push things. It turned out that when people didn’t know the ads were AI-generated, they preferred them over average human-made ads by about 10% to 20%. These were not top-tier agency ads, just average ads.
But when people found out the ads were AI-generated, credibility dropped and they preferred even the weaker human ads by 32%.
That’s the point I want to emphasize. I spoke with a client recently who was experimenting with Facebook ads and using AI-generated imagery. I told them that if people realize those images are fake, they could damage their brand.
We as an agency try to get authenticity into video. This isn’t self-promotion. We try to help clients be honest about what they’re good at and what they’re not good at. There’s a sales angle to this: people don’t want to be catfished. They don’t want the company that shows up to be different from what was promised.
Virginia Huling:
That goes back to solid Marketing 101. You are making a promise to your audience, and you need to fulfill it if you choose to make it.
I also think we are still in the discovery phase of what audiences will tolerate in terms of what’s real and what isn’t. For a lot of people, AI is still a toy. You see businesses following trends, using prompts, and posting AI images that all have the same look.
It will be interesting to see where the boundaries are. How much AI can you use to represent your business before people push back? Are you creating AI personalities that don’t exist and presenting them as real employees? Are people going to get tired of conflicting AI-generated news and turn away from it?
I think we’re still learning those boundaries. The creative industry is going to stay on the frontier of this because entertainment always pushes the envelope in the battle for attention.
David Maples:
That’s probably a good place to end this episode.
We covered a lot this week, and I think it’s important for you to understand how quickly the landscape is changing. My takeaway advice is to protect your foundation first. Build your frameworks agnostically so you’re not tied to one vendor or provider. If OpenAI jumps the price dramatically, you need options.
Keep your sensitive business data secure in a vault. Don’t put it into the machines if you can avoid it. Stay adaptable, because regulations are coming. As they roll out, tune into this podcast and we’ll try to cover them and give you a roadmap to navigate them.
Virginia Huling:
And remember: while the technology is changing rapidly, the core of your business remains human connection. AI is here to augment your imagination, not replace your relationships.
If you want to talk strategy or figure out how to implement these systems safely in your own business, and in a way that doesn’t alienate your team, reach out to us. We’re happy to help.
David Maples:
I really like that: “augment your imagination.”
At its core, this technology is almost like magic. There’s so much potential for destruction and disruption, and I guess magic is the same way.
There’s a famous line: “The future is already here. It’s just not evenly distributed yet.”
I do like “augment the imagination.” We need a tagline for the show. If you like the show, send us a comment and tell us what you think the tagline should be.
If you enjoyed this episode, hit like, subscribe, and share it with your professional network so we can keep these episodes coming and keep the conversation honest, human-focused, but technology-forward.
Virginia Huling:
For AI the Podcast, I’m Virginia Huling.
David Maples:
And I’m David Maples. Take care of each other out there.
Outro:
That’s a wrap for this episode of AI the Podcast. To make sure you never miss an update on the future of tech and business, hit that like and subscribe button wherever you’re listening right now. This show is made possible by the support of listeners like you. So if you want to help us grow this community, drop us a five-star rating. Thanks for listening, and we’ll catch you on the next one.