^{Just what we need.}

Three Hackers Used Claude to Break Into OpenAI In Less Than 72 Hours
https://gizmodo.com/three-hackers-used-claude-to-break-into-openai-in-less-than-72-hours-2000814009

Anthropic says its chatbot Claude is taking over the work of building its own successor - The Washington Post
https://www.washingtonpost.com/technology/2026/09/17/anthropic-says-its-chatbot-claude-is-taking-over-work-building-its-own-successor/

> Claude

Paywalled, but I believe they said last year or the beginning of this year that 90% of the code for the "next" model (i.e. the current one or perhaps the one before that) was written by Claude.

Anthropic specifically said that part of the reason they focused so much on coding was less that there was a huge market for it, but that once they got that right, it would dramatically accelerate all their other efforts. At the time, OpenAI was generally seen to be ahead, but Claude was betting that winning on coding would eventually result on winning overall, which seems to be the case.

>  Claude to Break Into OpenAI

That might get them to pay attention!