ryan_greenblatt comments on Sycophancy to subterfuge: Investigating reward tampering in large language models

[ryan_greenblatt]

Yep, reasonable summary.

I do think that it’s important to keep in mind that for threat models where reward hacking is reinforced, small RTFs might matter a lot, but maybe that was already obvious to everyone.

I don’t think this was obvious to everyone and I appreciate this point—I edited my earlier comment to more explicitly note my agreement.