Will Anthropic be the best on AI safety among major AI labs at the end of 2025?
Mini
12
Ṁ488
2026
87%
chance

Resolves NO if on 31 December 2025 another major AI lab is, all things considered, better than Anthropic on overall AI safety. Otherwise resolves YES.

This question will defer to ratings by AI Lab Watch, or if for some reason that website no longer updates or becomes unreliable, my own reasonable judgement using similar criteria.

Get Ṁ1,000 play money
Sort by:

the ai lab watch metrics seem super sus

This all depends on whether Ilya's new lab gets included

I think Ilya's new lab is hard to evaluate until they put out any kind of research.