david rein(@idavidrein):Some important context here: I understand the "20% of compute" claim in this blog post to be saying that monitoring overhead is 20% of the inference compute being monitored. The claim is not "20% of compute at OpenAI is now being used for monitoring"

david rein

@idavidrein

Leading embedded stress-testing @METR_Evals. Formerly: early employee @cohere, made GPQA @nyuniversity

Joined July 2018

1.4K Following 5.6K Followers

david rein@idavidrein

2026.08.19 18:41

Some important context here: I understand the "20% of compute" claim in this blog post to be saying that monitoring overhead is 20% of the inference compute being monitored. The claim is not "20% of compute at OpenAI is now being used for monitoring"