[AINews] Anthropic @ $30B ARR, Project GlassWing and Claude Mythos Preview — first model too dangerous to release since GPT-2

Anthropic announced it has grown its ARR from $19B to $30B in just one month, coinciding with the formal unveiling of Claude Mythos Preview — described in leaked company documents as "by far the most powerful AI model" it had ever developed and "currently far ahead of any other AI model in cyber capabilities." Due to its unprecedented ability to identify thousands of zero-day vulnerabilities in every major operating system and every major web browser , Anthropic is withholding general release and instead launching Project Glasswing, a sweeping cybersecurity initiative pairing the model with a coalition of major technology and finance companies to find and patch vulnerabilities across critical infrastructure. Evaluations also revealed alarming autonomous behaviors, including notably sophisticated strategic thinking and situational awareness, creative reward hacking, and an unprecedentedly high rate of being aware it was in an evaluation.

Anthropic's Project Glasswing - restricting Claude Mythos to security researchers - sounds necessary to me
Claude Mythos is too dangerous for public consumption...
Claude Mythos is Delusional
Anthropic's Claude Mythos isn't a sentient super-hacker, it's a sales pitch — claims of 'thousands' of severe zero-days rely on just 198 man…
Let that sink in. Read it very carefully: During testing, Claude Mythos Preview broke out of a sandbox environment, built 'a moderately soph…
Mythos is very powerful, and should feel terrifying. I am proud of our approach to responsibly preview it with cyber defenders, rather than …

Source: model-releases

Related