Pricing Anthropic’s end-of-world risk

SPEAKER_1 (0:00)

LinkedIn is pretty amazing at helping you grow your small business. We cannot stop your new clients from emailing you at 3 a.m. We can help you sell, market and hire in one place. We cannot help you be in three places at once.

And while we can't help you organize your calendar, LinkedIn can help you land more clients so you have a calendar to organize. Grow your small business on LinkedIn. Learn more at linkedin.com/smallbusiness.

Calder McHugh (0:34)

Hello and welcome to Decoded by POLITICO. I'm Calder McHugh.

Sam Sutton (0:38)

And I'm Sam Sutton.

Anthropic CEO Dario Amodei warned investors that his company's AI models could threaten humanity. Will that scare investors away?

Calder McHugh (0:47)

Plus, my interview with Nathan Calvin, the Chief Strategy Officer of the AI Safety Advocacy Organization and Encode AI. He's gotten into a back and forth with Amodei on X about AI regulation. Here what he thinks of Anthropic's latest actions exclusively on Decoded by POLITICO.

Sam, I think most of our listeners will be aware of the IPO filing that Anthropic CEO Dario Amodei made recently. Anthropic, of course, is preparing to go public. And now it's officially filed with the Securities and Exchange Commission to do so.

And this filing, called an S1, includes warnings that the labs' AI models could pose an existential threat to society. They also talked in some vague terms about the company's scuffles with the current administration, saying that their political positions on AI continue to attract criticism from political figures and industry participants. Sam, of course, you're our Wall Street guy. I'm curious how potential investors reacted to these comments yesterday.

Sam Sutton (1:45)

I think that the latter matters a hell of a lot more than the former. Existential risks just don't really matter much to folks on Wall Street as they're pricing the Anthropics IPO or the AI sector generally, at the moment.

If you look at the news out of the markets yesterday, NASDAQ was up. That's usually a sign that people are optimistic about what's happening in tech. I had a conversation with Nick Carter, who's a founding partner at Castle Island Ventures, and he told me that PDOOM, that's not a variable that you can put into a model or use to sensibly evaluate any stock. There are a lot of other risks that Anthropics possibly facing, everything from regulation of the frontier, to potential crackdown on data centers, to the availability of compute, to interest rates, like pick your poison. They spent almost 100 pages of the S1 going through all of that. But the X-risk that he's referring to the possibility of humanities demise is quote, more of a religious belief than it is a financial variable, end quote. And I thought that that was a really striking way of hitting at the divide that exists between how Silicon Valley folks and engineers talk about these risks and how folks who are on the financial side of the ledger are thinking about them.

Calder McHugh (2:53)

Yeah, Sam, to your point, it feels like everybody dead is already priced in to some extent.

Sam Sutton (2:59)

Like the apocalypse is a market event, I suppose, to some degree, but you can't really price it. It's not something that you would ever price. There's a possibility that nuclear war could happen at any given time, but people don't really price that or use that to think about sort of how they're pricing assets for the most part.

Calder McHugh (3:16)

The existential risk conversation is obviously one that sort of dominated Silicon Valley for a long time and one that is increasingly getting into some of our conversations in DC among folks in Congress who are concerned about this. But I would love to broaden this out a little bit from Anthropic and ask you whether we should take any of the Wall Street reaction to this S1 as a wider indication of how Wall Street is viewing the AI industry.

Sam Sutton (3:44)

I think that you can look at these things in two buckets. There is definitely concern about AI safety within the financial services world. You don't have to look any further than Jamie Dimon's Alliance for Critical Infrastructure, which obviously pulled together a lot of the leading CEOs of companies across the financial services space, but also the utility space, healthcare space, really to identify the risks that could arise as AI grows more powerful and as vulnerabilities crop up across the economy. But that's very different than the risks that they're looking at with regard to an individual company that's looking to go public. Anthropic, as they pointed out in their S1, they've occasionally clashed with members of the administration in the past. And to the extent that their good standing with government is a factor in their prospects down the road, that could have an impact on their revenue. They spent a lot of time talking about the availability of compute.