Claude Mythos market scare
Coverage this week linked fears about Anthropic's Claude Mythos model to a roughly $2 trillion wipeout in IT stocks and quoted warnings that AI could worsen certain risks. (m.economictimes.com)
Anthropic’s new Claude Mythos model sharpened a selloff in United States software stocks this week as investors priced in faster artificial intelligence disruption. (reuters.com) Reuters reported on April 9 that United States software shares fell after Anthropic’s update revived fears that artificial intelligence tools could automate more white-collar software work. Bloomberg had already described a broader February rout that erased nearly $1 trillion from the value of software stocks tracked by an iShares exchange-traded fund in seven days. (reuters.com) (bloomberg.com) Anthropic announced Claude Mythos Preview on April 7 and said it would not make the model generally available. The company said the system was strong enough at finding and exploiting software flaws that it would limit access to selected partners in a defensive program called Project Glasswing. (anthropic.com) (cnbc.com) A software vulnerability is a bug that can be turned into a break-in, like a bad lock on a server or browser. Anthropic said Mythos had already found thousands of high-severity vulnerabilities, including some in every major operating system and web browser. (anthropic.com) Project Glasswing includes Amazon Web Services, Apple, Broadcom, Cisco, CrowdStrike, Google, JPMorganChase, the Linux Foundation, Microsoft, NVIDIA and Palo Alto Networks. Anthropic said it is committing up to $100 million in usage credits and $4 million in donations to open-source security groups. (anthropic.com) Anthropic’s own system card says Mythos is its “most capable frontier model to date” and shows a “striking leap” over Claude Opus 4.6 on benchmarks. The same document says that increase in capability led Anthropic to withhold a general release and keep the model inside a limited cybersecurity program. (anthropic.com) The scare did not start with the April 7 launch. Fortune reviewed draft Anthropic materials in a publicly searchable data cache in late March, and CNBC said cybersecurity stocks fell after that report highlighted the model’s offensive cyber capabilities. (finance.yahoo.com) (cnbc.com) Jamie Dimon’s annual shareholder letter added to the week’s risk backdrop, though his warning was broader than Anthropic alone. In the April 6 letter, the JPMorganChase chief executive cited risks from geopolitics, inflation, private markets and artificial intelligence. (cnbc.com) Anthropic and its partners are framing Mythos as a defensive tool for patching critical software before attackers get there first. Markets treated the same capability as evidence that artificial intelligence is moving from a product story to a pricing threat for software companies. (anthropic.com) (reuters.com)
