1 paper · 1 filter
Matthew Riemer, Tommaso Tosato, Amin Memarian +4
This position paper argues that AI agents with chain-of-thought reasoning capabilities are predisposed to exhibit collusive behavior and should be required to obtain behavioral cer…