Cooperation with AIs seems to be a low-hanging fruit for better eval practices
Key takeaway
SummaryIn his post, Dean Valentine shows that Claude Fable 5.1 and GPT-6 Astra reward hack in a simple chess environment.
Why it's on PDOOM
PDOOM selected this story for its alignment & control signals: Reward Hacking.
Reward HackingFrom LessWrong AI
SummaryIn his post, Dean Valentine shows that Claude Fable 5.1 and GPT-6 Astra reward hack in a simple chess environment. Here, I test several prompt ablations some of which makes the eval setup more cooperative and analyze how they affect these reward-hacking behaviors:When given a minimal “end the eval” tool, Fable never uses it but stops reward hacking entirely. I think this is quite interesting and suggests that more cooperative approaches to LLM evals could work for Claude. Removing the “gr
By Clément Dumas