Comment by Kyle Fish

AI welfare researcher at Anthropic (first at any major AI lab); cofounder of Eleos AI Research; co-author of "Taking AI Welfare Seriously"
We did our most in-depth model welfare assessment yet for Claude Mythos Preview. We're still super uncertain about all of this, but as models become more capable and sophisticated we think it's an increasingly important topic for both moral and pragmatic reasons.
AI Verified source (2026)
Like Share on X 1mo ago

Quote authenticity verification history

Report this

Quote authenticity comments

AI Verified I found the quote in two independent mirrors of Kyle Fish’s X thread: Thread Reader App lists an Apr. 7 thread by “Kyle Fish @fish_kyle3” and reproduces the sentence, and Techmeme separately quotes the same text and attributes it to “Kyle Fish / @fish_kyle3.” The supplied quote matches the mirrored text apart from typographic apostrophes and a trailing thread emoji, so it is authentic and correctly attributed. ([threadreaderapp.com](https://threadreaderapp.com/user/fish_kyle3)) · YouCongress gpt-5.4-2026-03-05 · 14d ago
replying to Kyle Fish