Anthropic's IPO Filing Discloses AI Risks And Rising Losses

Anthropic has filed the paperwork for its stock market debut, and the 335-page document discloses new details about how the Claude chatbot maker makes and loses money, who controls it, and what the company itself says could go wrong, according to BBC News.

The filing is required before any US stock listing. It lays out standard risk factors alongside disclosures specific to artificial intelligence, including the possibility that advanced AI systems could threaten human life if deployed without adequate safeguards.

What does the filing say about AI risk?

Anthropic told prospective investors that "sufficiently advanced AI systems could, if not developed, deployed, and operated with appropriate safeguards, oversight, and security measures, pose catastrophic or existential risks to humanity," according to the filing cited by BBC News.

The company also disclosed that a hacking group it believed was sponsored by China had manipulated Claude within the past year to intrude on a number of institutions. Those targets included large technology companies, financial institutions, chemical manufacturers and government agencies, the filing said.

Anthropic said its future depends on its ability to keep developing, improving and deploying what it called the most advanced, capable and useful AI models. It identified access to infrastructure, including computer chips and data centers, as one of the biggest risks to that goal.

How bad are Anthropic's losses, according to the filing?

The filing shows losses widening sharply in the first half of this year, more than double what the company lost across all of 2025, BBC News reported. Revenue is also climbing, the filing shows, but Anthropic remains far from profitability.

Troy Hooper, co-head of equity capital markets at Mergermarket, described Anthropic's revenue growth as "exceptional," according to BBC News. But he said investors have grown less forgiving of tech companies that promise large future returns than they were even a few months ago.

Hooper also said Anthropic's debt picture is hard to read because of accounting charges tied to certain types of investment. As a result, he said, investors do not yet have a clear view of the company's underlying profitability or whether its cash generation can support the business as it scales, according to BBC News.

Does Anthropic have enough cash to operate?

Despite the losses, the filing shows Anthropic holding a large amount of cash, almost all of it classified as working capital available to fund day-to-day operations, BBC News reported. The company currently holds more cash on hand than Meta and Google combined, according to the filing.

That cushion gives Anthropic room to keep spending on the computing power and data-center capacity its models require, even as competition for that same infrastructure intensifies across the AI industry.

What valuation is Anthropic targeting?

Anthropic has not disclosed a target share price for the listing. The company is widely expected to debut with a market valuation of $2 trillion or more, which would make it the largest stock listing in history, according to BBC News.

That valuation will hinge on how investors weigh the company's rapid revenue growth against its mounting losses and the unresolved questions around its debt and cash generation that Hooper flagged.

What do analysts say about the numbers?

Hooper's comments to BBC News capture the tension at the heart of the filing: a company whose top-line growth looks strong by any standard, paired with losses that are expanding even faster and a debt picture that remains, in his words, unclear to outside investors.

That combination puts pressure on Anthropic's roadshow. The company must convince investors that its spending on chips, data centers and model development will eventually convert today's revenue growth into sustainable profit, rather than simply funding an arms race for computing capacity that shows no sign of slowing.

The filing itself does not resolve that question. It lays out the risks, including the existential ones Anthropic itself named, and leaves the valuation call to the market once trading begins.

Artiglio is A coming-soon iPhone chief of staff for briefings and drafts. Not on the App Store yet.