Are The AI Labs Getting Nationalized?

Bullish:

• $VISA - Visa has locked-in distribution and brand ubiquity that AI cannot easily disrupt. The company can use AI to dramatically reduce costs while maintaining pricing power and revenue, making it an attractive investment despite appearing vulnerable to disruption.

  • “you take a company like Visa, which at face value should be, like, the easiest to disrupt company in the world. Right? Crypto, we you and I have seen a million pitch decks saying crypto is gonna disrupt Visa. AI should disrupt Visa. And yet, maybe not because, Visa can use stablecoins. Visa can use AI to drive its cost down dramatically while providing the same service. And why is Visa gonna get replaced? So, yeah, OpenAI could offer Visa with, whatever, half the credit card payment processing fees, but does that really matter to consumers? Does that really matter to merchants compared to the brand ubiquity of it being, accepted everywhere around the world, all that? So his argument is that by reducing costs for companies with locked in brands, consumer bases, very high switching costs, all of that, those companies are gonna see falling costs due to AI and flat or growing revenue, which makes them great buys.”

    Ari·41:47
• $BTC - Bitcoin has strong secular tailwinds and prices are likely to continue going up, though the easiest money has been made. The asset still has a strong tailwind despite being past the early adoption phase.

  • “I think crypto is, like, maybe where I'm sorry. AI is maybe where crypto was in 2021 or 2022, which is to say the easiest money is gone. The the tailwind that was so strong that almost anything with an AI name gave you a 100 x. I think that's over. Now this is more like '21, '22, which is to say we still have a strong tailwind. Prices in general are still gonna go up.”

    Ari·38:31
• $BIOTECH - Biotech companies are positioned to benefit from rapid advancements in research and AI-aided drug discovery. The industry is at the precipice of radical change for human health with an overwhelming amount of new drugs coming to market.

  • “this entire conversation makes me so bullish on the idea of these biotech companies. I mean, there's so much that's coming out right now in the research and in the literature and also aided by drug discovery using AI. I mean, Eli Lilly just kind all this cholesterol pill. I mean, we're kind of at the precipice of radical change for for human health.”

    Avi·54:10
• $INFRA - Infrastructure companies with locked-in distribution and regulatory moats can benefit from AI by reducing costs while maintaining pricing power. These boring old-world companies with high switching costs are attractive investments in the AI era.

  • “there's a hedge fund manager, I think it's Chris Hahn at Founders Fund, who's one of the top performing managers of all time. And he has a, he's had a kind of a interesting thesis that's a bit of a tweak on Warren Buffett's kind of value investing where he invests in companies with locked in distribution. Sometimes those are government monopolies or regulatory capture. Sometimes those are literal distribution pipelines that would be very expensive to replicate. And he kind of doubled down on that thesis in the era of AI, and he's been doing very well over the last few years. And his thesis is that an AI company can't easily disrupt a distribution company.”

    Ari·40:08

Bearish:

• $OPENAI - OpenAI faces significant risks from corporate and nation-state espionage, regulatory nationalization concerns, and value capture challenges. The most valuable models may be run privately by insiders rather than monetized for shareholders, and government control of AI engineers could impact the business.

  • “I've been I don't know if bearish is the right word, but a little bit skeptical on the IP created of the AI generators. So, like, Claude, Meta, Google. I was having a debate with a a close friend who's pretty senior on in the Meta ML team about this, and we we were debating it. It was, like, forty eight hours before Claude accidentally posted a good chunk of their code base publicly. This was, like, three months ago. They accidentally revealed a lot of weights and things. There's constant corporate corporate espionage both between the firms. Right? Google's trying to poach top Meta people and vice versa, but then nation state espionage. Anything being produced by Claude, by Meta, by Google, you have to assume is in the hands of Russia, China, North Korea.”

    Ari·24:03

  • “If a firm like OpenAI or Claude develops a model that can, for example, beat the stock market tremendously, Not only are their users probably never getting access to that model, so we as the public are never getting access to that model. That's not gonna be on, like, a like, a retail customer tier list. Right? The enterprise users are also not getting access. Ken Griffin's not getting access to that, but neither are the shareholders. Because if Sam Altman realizes what he has, why is OpenAI gonna run that out of? Sam Altman's gonna run it privately, or the engineer who discovered this is gonna run it privately.”

    Ari·25:13

  • “AI is that. Mythos is a military project. And at some point, the US government and other governments I mean, governments are already doing this, these Chinese, US government may start approaching it that way and say, oh, you're a top engineer on, you know, machine learning at Meta. You're only using a government issued cell phone. You're only traveling with permission. You're not allowed to leave the country. You're not you know? And if if US doesn't do that, we will be leapfrogged by China because they'll they'll they they can basically use corporate espionage to get to where we are and then and then go forward. So all of that affects these as investments, right, in ways I don't fully understand.”

    Ari·0:26
• $ANTHROPIC - Anthropic faces similar risks to OpenAI including IP theft, potential government nationalization, and restrictions on selling or using models. The company may not be able to capture the full value of its technology due to regulatory and national security concerns.

  • “my impression talking to other investors was that they were not discounting this at all. They they were not incorporating these factors as risks, either the nationalization of Entropic that, know, Entropic not being allowed to sell or use mythos.”

    Ari·27:13
• $DATACENTER - Data center buildout may be analogous to fiber optic overbuilding in the 1990s. While demand exists, other bottlenecks (power, rare earth metals, algorithmic efficiency) are emerging, and recent AI papers are dramatically cutting hardware needs, suggesting overinvestment in compute infrastructure.

  • “I worry that the buildup in data centers is and this is not other people accept this for sure is analogous to the overbuilding of fiber in the Internet era in the mid to late nineties. It's not that demand for the Internet didn't continue growing exponentially. It's that other bottlenecks appear that had to be addressed first. And so with AI, what we found is data centers were the bottleneck. We threw massive, unbelievable amounts of money at that and built a crazy it was basically like a man on the moon project. Like, building a trillion dollars of data centers in a couple years is incredible. But now what we're realizing is there's other bottlenecks. Yes. We could get better AI if we threw another 5,000,000,000,000 at compute, but everyone's realizing that's not the good ROI investment at this point. And so what we're starting to see is week after week, new AI papers being published that dramatically cut the hardware needs to achieve the same result.”

    Ari·29:41
• $ACCENTURE - AI and LLMs are disrupting consulting and analyst work, with one person now able to do the work of multiple junior analysts. This suggests structural headwinds for traditional consulting firms that rely on large teams of junior staff.

  • “If I was launching a hedge fund tomorrow, I don't think I would hire junior analysts. I think I would just use AI in their place. Basically, if I I mean, Blocktower, we were very thinly staffed because I kinda do, like, math and people. So we were we were always understaffed. We were at one point, we were managing almost $2,000,000,000 with a team of, like, 33 people across, like, five different strategies and franchises. But if I did it over again, I think I'd probably have a third the head count. Because most junior analysts, I think it would be basically me and an LLM and or or I would hire a AI power using single analyst, and he would do the work of eight analysts”

    Ari·43:21