Anthropic investigating unauthorised access of powerful Mythos AI model

This is a puff piece designed to build excitement around Anthropic's products.

This strategy of making unverifiable claims to possess a super-weapon is one of the most dystopian marketing initiatives ever. Makes Fritz Haber look like a deeply-committed humanitarian. Wild.

The claims are verifiable and the Firefox team published one verification just yesterday (https://blog.mozilla.org/en/firefox/ai-security-zero-day-vul...).

Open source project dependent on VC funding models verifies claims of VC funded corporation.

Perhaps, but it also introduced a lot of low hanging fruit about the security übermodel not being secured.

If this builds excitement for Mythos then we really are living in the weirdest of timelines.

https://archive.ph/7v9WF

I tire of AI writing marketing puff peices for AI companies which are mostly read by AI. It's like an ouroboros of slop.