US regulators scrutinise Anthropic’s Mythos as market anticipation for IPO wanes

Mounting concerns over Anthropic’s latest AI model, Mythos, have shifted the focus from investor optimism to regulatory risks, with market valuations and official warnings highlighting the growing ...

The White House and Treasury Department’s growing concern over Anthropic’s latest AI system has spilled into the markets, with a thinly traded Polymarket contract on the company’s eventual IPO valuation slipping to 0.6% for a debut market capitalisation between $100bn and $200bn by the end of 2027. The move came after Treasury Secretary Scott Bessent and Federal Reserve Chair Jerome Powell met major US bank chiefs in Washington to discuss the cybersecurity risks tied to Claude Mythos, Anthropic’s newest model, which researchers say can uncover thousands of software vulnerabilities.

The market has become increasingly sensitive to signs that regulatory scrutiny could complicate any listing. The contract had been closer to 1% the day before, but with only a small amount of daily USDC changing hands, prices are highly exposed to modest orders. In such a shallow market, a relatively small bet can move the odds sharply, leaving the contract vulnerable to abrupt swings if more negative news emerges.

The Treasury’s interest has also sharpened the sense that Mythos has become a policy issue, not just a technical one. Bloomberg reported that the department’s technology team is seeking access to the model to probe for flaws, while The Guardian said the April 10 meeting in Washington was convened to assess the security implications for the banking sector. That follows reports that Anthropic has already discussed the model with White House officials, suggesting the company is now navigating both commercial enthusiasm and official caution.

The wider backdrop is one of extraordinary investor optimism. Tom’s Hardware reported on April 23 that Anthropic’s private secondary-market valuation had climbed to about $1tn, overtaking OpenAI’s $880bn estimate in some trading, even as security concerns around Mythos deepen. That contrast highlights the market’s split view: investors continue to price Anthropic as one of the most valuable private AI companies in the world, while policymakers and banks are focusing on the possibility that its most advanced model could create new cyber risks before it ever reaches the public market.

Source Reference Map Inspired by headline at: [1]

Sources by paragraph: - Paragraph 1: [2], [3] - Paragraph 2: [1] - Paragraph 3: [2], [3], [4] - Paragraph 4: [5], [6], [7]

Source: Noah Wire Services