For Claude, capability and dispreferring CDT are the ~same thing. Much more so than for GPT.
We've previously reported that decision-theoretic capabilities and favoring EDT/generalised-one-boxing over CDT correlate in LLMs (both measured by DTBench). (Note that EDT, for the most part, doesn't come apart from FDT / UDT on DTBench.[1]) Anthropic also replicate the same finding in their Opus 4.7 and Fable 5 model cards.We recently noticed something funny: Capabilities and preference against CDT answers basically perfectly for Anthropic models. This holds whether you measure capabilities us
By Chi Nguyen