Half the country is using AI daily, yet nearly everyone is convinced it's going to ruin society. Welcome to the paradox of 2026. It's absolutely wild. We're outsourcing our daily emails to models we secretly fear. Welcome to the deep dive. Today we're unpacking the G7 summit's shift toward government AI equity grabs. Right, and we'll run through a rapid -fire update of the newest models hitting the market. Plus, we'll explore that massive disconnect between AI adoption and
public trust. Now, neutrally speaking, there are several different political proposals right now. The Treasury Secretary wants to use AI equity to fund Trump accounts. Right. But the Commerce Secretary prefers routing equity into a sovereign wealth fund. And Senator Bernie Sanders proposed a totally different law. He wants to give Americans direct ownership stakes in top firms. So the source facts show different paths to the exact same idea. The public sector wants financial
value directly from private labs. Exactly. The specific political flavor just determines where the money actually goes. The timing of this is incredibly suspect to me, though. Oh, totally. These talks happened right before the fable and mythos restrictions hit. You know, that massive export control clash. with Anthropic. Yeah, Anthropic's Dario Amadei and Google DeepMind's Demis Hassabis were pushing hard. They want a robust U .S.-led
coalition on AI rules and chips. Right. They're trying to shape the cage before they get locked inside. Exactly. The industry's best case scenario is getting Fable 5 back online. And they want to do it without aggressive power grabs. I mean, there's also talk of naturally slowing down the release cadence. This whole new model every month schedule is just fundamentally broken. It exhausts the developers entirely. And it severely confuses everyday users trying to keep up. I completely
agree with that exhaustion. I mean, I still wrestle with prompt drift myself. Oh, yeah. We all feel that pain. You know the exact feeling. Your go -to instructions, just quietly stop working under the hood. Yeah, it breaks the user experience completely. But zooming out, we're talking about frontier models here. Let's define that real quick. Frontier models are the most advanced, cutting -edge AI systems currently available. Right. And the worst -case scenario from the
sources is pretty dark. Imagine frontier models locked entire... behind massive government gates. Oh, wow. Yeah. Open source development gets banned completely for national security reasons. So the market just faces zero actual competition. Exactly. We end up paying whatever the big labs demand. Or we settle for vastly subpar intelligence in our daily tools. Are we heading toward a world where AI is treated as a public utility or just
a government sanctioned monopoly? It looks like they want the broad structure of a public utility, but the control remains tightly centralized among a few key political players. Public utility vibes, but entirely controlled by government gatekeepers. Beat. Precisely. And that central control is what labs are fighting. The government clearly wants to slow things down right now. Right. But the tech labs are sprinting in the absolute opposite direction. They really are moving faster than
ever before. GPT 5 .6 and the pro version might actually arrive next week. Yeah, and early testing shows their logical understanding has improved massively. But the coding limitations are still a stubborn bottleneck. Yeah, front -end and web dev are definitely not fully solved yet. No, they still struggle with complex architectural planning. But Anthropix's Chris Ciari is very confident about CloudFable 5. He believes the models will return to the market very soon. And
the tool updates are just as aggressive. Perplexity Computer just received a massive brain upgrade for max users. It literally makes you want to provide more content. text every day. It learns your specific workflows perfectly. Google's pushing hard into the enterprise ad space too, right? Yeah. Google's new Ask Ad Manager agent can troubleshoot complex marketing campaigns. But we're also seeing some fascinating cultural shifts around this technology. Match Group recently released some
wild data about Tinder. Oh, right. 47 % of U .S. singles feel negative about AI in dating. Which isn't surprising. You don't want a robot writing your romance. Right, but look at the massive contradiction in their behavior. Many of those same singles still want help with their profiles. They hate the concept, but they love the practical utility. Exactly. They still want AI to help navigate awkward intro chats. On the corporate side, companies are getting very creative
with structure. Snap is spinning off its internal AI video team as Dotmo. Which just sounds like basic corporate cost cutting on the surface. But Snap is keeping a large equity bet in the spinoff. Right. They want the upside. Without the massive operational burn rate. Right. And that burn rate is driven entirely by the hardware bottleneck. Absolutely. Amazon is apparently considering selling AI chips to outside companies. A challenge NVIDIA. Yeah, they're trying to carve
out their own massive turf. AWS CEO Andy Jassy estimates a $50 billion chip run rate. $50 billion? Beat. That's a staggering amount of hardware. It's almost impossible to comprehend that scale. Whoa, imagine scaling to a billion queries. Two secs silence. It requires a fundamental rewiring of our global energy grid. And that compute power fuels incredibly ambitious new startups today. Like Odyssey, right? They just raised $310 million. Yeah, at a $1 .45 billion valuation. And they're
building world models. Basically, AI that simulates real world physics and complex human behavior. Exactly. It's rendering a universe, not just writing an essay. So with world models perfectly simulating physical reality, is the hardware demand ever going to plateau? The physical world has infinite resolution and infinite complexity. Simulating it perfectly requires exponentially growing processing power. It definitely won't stop anytime soon. Not until the physical and
digital worlds completely merge. That's the ultimate endgame for these massive frontier labs. But let's bring this back down to the human scale. Yeah, let's do that. That massive hardware bottleneck is exactly why everyday users feel squeezed. Which brings us to the microtools patching specific daily gaps. These microtools are solving specific friction points for regular workers. Seaton's 2 .0 Mini is a fantastic example of this trend. Yeah, it generates high -quality AI videos in
under a single minute. Faster, cheaper, and they offer 30 free credits. It completely democratizes video generation for the average creator. And then there's Upstream, completely changing the daily administrative grind. It sorts your messages and drafts your replies automatically. It makes checking your email feel light, fast, and fun again. Honestly is another tool providing massive value to modern digital marketers. It pulls sentiment
insights directly from Reddit. TikTok and X. It also scrapes data from YouTube, Instagram, and Facebook automatically. You get unfiltered feedback without spending hours reading angry comments manually. I want to highlight Tabstack specifically for the developers listening today. Tabstack is a highly specialized web data and automation API. You pass a URL and it spits back pristine JSON. Which is just a simple text format
for storing and moving digital data. It delivers highly reliable structured output from a single URL. It completely eliminates the parsing errors that plague legacy web scrapers. Are these micro tools permanently replacing whole software suites or just bridging temporary gaps? Right now, they act as brilliant little patches. But as they interlock seamlessly, they will obsolete massive legacy platforms entirely. They're bridging small gaps today, replacing entire suites tomorrow.
Beat. Yeah. So let's recenter on the actual human beings using these platforms. Yeah, that's arguably the most important question we face today. We've covered massive models and the incredible utility of microtools. But how do we actually feel about integrating this technology? Well, Pew Research just dropped its massive 2026 data report. They surveyed over 5000 U .S. adults. And the findings highlight a very stark and deeply troubling contrast.
Yeah. Adoption rates are climbing rapidly everywhere. Roughly half of all U .S. adults are now using AI chat bots. And a solid 25 percent log in every single day. But public pessimism is heavily outpacing the optimism right now. Nearly 40 percent expect AI to make. our society fundamentally worse. Exactly. They believe it will degrade society over the next two decades. Only 16 % actually believe it will make things better. We're rapidly adopting a technology we fundamentally fear right
now. And there's a fascinating demographic quirk hidden inside this survey data. We really need to point out the massive under 30 divide here. Oh, right. Younger users lean on AI the absolute hardest. But they're also the absolute biggest skeptics in the entire survey. Just 14 % of them see a positive long -term payoff. It feels so counterintuitive at first glance. They grew up entirely immersed in modern digital technology. They see the potential for severe societal disruption
very clearly. Let's look at the actual market share reality check here. ChatGPT is still the undisputed heavyweight at 44 % usage. Yeah, and Google's Gemini currently holds about 24 % of the market. But Anthropic's Claude sits way down at just 6%. Despite being constantly discussed in tech circles, Claude barely registers. It absolutely pales in comparison to the massive footprint of chat GPT. Why do we keep using these tools daily if we genuinely believe they'll eventually
ruin society? The immediate benefit to your personal productivity is highly visible right now. That tangible benefit completely overrides abstract fears about long -term decay. Because personal immediate convenience almost always overrides long -term societal anxiety. Exactly. And if we look at the core narrative thread of this deep dive, we're trapped in a massive three -way tug of war. Tech labs are moving at breakneck
speed to simulate reality perfectly. Right. And the government is frantically scrambling to lock down control. They literally want the deed to the technological foundation. And the broader public is caught squarely in the middle. We're relying heavily on these tools to survive the modern workday. But we fundamentally distrust where this technological revolution is actually heading. I want to leave you with a final provocative thought today. Something for you to deeply mull
over on your own time. Let's hear it. We discussed the government securing equity in these massive AI labs. Imagine they eventually distribute those financial dividends directly to regular citizens. Like a universal basic income. Exactly. If that happens, will that pay out finally by the public's trust? Or will it just deepen our skepticism of the machines running our lives? Thank you for taking this deep dive with us. Otero Music.