“Anthropic on Monday rolled out the prospectus for its long-awaited initial public offering, and reading all the financial, regulatory and legal risks makes us wonder if it's trying to scare investors.
There's never been a public IPO like Anthropic, with such a promising but also risky technology. The paradox is laced through its more than 300-page prospectus as it repeatedly says it can benefit humanity and solve humanity's biggest problems. Yet it also says AI may pose "catastrophic or existential risks to humanity."
This has been CEO Dario Amodei's public contradiction, and now that he's selling shares perhaps he could make up his mind. As a friend quips, perhaps he should ask Claude.
---
Beyond this small existential matter, there are other risks. The largest is whether Anthropic will ever generate the revenue to cover its soaring costs to stay at the AI frontier. Anthropic reported $16.3 billion in revenue in the first half of this year, up from $1.2 billion in the same period last year.
Anthropic also says its customers include more than half of the Fortune 500 companies, including the 10 largest internet firms, nine of the 10 largest pharma companies, the 10 largest banks, nine of the 10 largest semiconductor companies and eight of the 10 largest healthcare firms.
This is all impressive, yet it says a "significant portion of our customer base consists of startups and technology-oriented enterprises that may be more sensitive to adverse macroeconomic conditions." Beware a major economic slowdown.
The prospectus names Google and OpenAI as chief competitors on the frontier, while open-weight and open-source AI models "may capture a portion of the market for certain use cases, which could exert pricing pressure on our offerings or lead certain customers to adopt open-source alternatives for those use cases."
Large banks and infrastructure providers may be reluctant to use Chinese models because of potential security risks. But Nvidia and Meta are pushing their own open-weight models, which could be attractive because of lower costs. Companies needn't worry about sharing proprietary data if they use open models.
A recent report found that open models account for nearly 60% of AI use in the U.S., up from 10% a year ago. If open-weight models proliferate, Anthropic could struggle to pay for its computing build-out. Anthropic estimates it has a minimum of $417 billion in future payments, mainly owed for computing and infrastructure.
Then there are regulatory risks. The prospectus says that "recurring narratives concerning the potential adverse societal impacts of AI, including that AI may displace or eliminate jobs, may be misused or otherwise cause harm, and may pose catastrophic or existential risks to humanity" could hurt its reputation and put off potential customers.
No mention that Mr. Amodei is a leading promoter of such narratives. The political risk also isn't trivial. In June the Commerce Department imposed worldwide export restrictions on Anthropic's Fable 5 and Mythos 5 models, which in effect forced the company to disable them.
The government restrictions were later lifted, but "similar export controls could recur, could remain in place longer, or could extend to additional models or capabilities," the company warns. It omits that the risk of Washington expanding political control over AI will increase if Democrats win the White House in 2028.
In an understatement of the modern age, Anthropic says "we are subject to complex and evolving laws, regulations, and other legal obligations related to privacy and data protection." Chinese AI labs face few of these barriers.
Becoming a public company will increase Anthropic's legal risks. By now Mr. Amodei must know that anything he says can and will be used against him by regulators and the plaintiff bar. If an Anthropic agent causes damage, the tort and securities bars will pounce.
On Friday the owners of Elvis Presley's music sued Anthropic, alleging it trained its models on more than 200 of his songs without permission. They are seeking more than $41 million in damages. "Costs of defending against such litigation, regardless of merit, are considerable," the prospectus says. Yes they are.
---
Given such enormous risks, why is Anthropic going public now? Our guess is it's trying to take advantage of the AI boom to let employees with stock options and early investors cash out before its edge erodes -- or before more risks bite.
One underplayed hint of potential trouble is Anthropic's insistence on its "public benefit purpose," which includes ensuring "the hardest problems facing humanity are not left to market forces alone to solve." That's more or less an admission that Anthropic will continue advocating for a bigger government role in controlling AI. Investors and taxpayers, take heed.” [1]
Let us hope that China’s open weights AI will defeat our greed and clumsiness, realizing AI promise to the world.
1. Anthropic's Big IPO Gamble. Wall Street Journal, Eastern edition; New York, N.Y.. 27 Oct 2026: A14.
Komentarų nėra:
Rašyti komentarą