FDT Defenders Push Back on "Logical Dependency Hypotheses Aren't Really Logical"
jessi_cata · x · 2026-11-10
A deep debate over decision theory. Someone challenged whether the "dependency hypotheses" in the FDT (functional decision theory) family are "logical" in any meaningful sense. jessicata countered: the key point is that the causal graph contains logical nodes, and interventions act precisely on those nodes — the graph structure is meant to express counterfactual dependence on logical facts, whereas a CDT graph can exclude logical nodes entirely. The author also noted an asymmetry: physical CDT has a constructible form (e.g., wiring a quantum random number generator into an RL exploration algorithm), while FDT has no equivalent constructible form yet.
More from Safety
- Guidelight argues OpenAI's Control Room doesn't solve AI interfering with its own monitors — sjgadler · 2026-11-10
- A concise summary of the core argument in 'If Anyone Builds It, Everyone Dies' — DavidSKrueger · 2026-11-10
- Circuit breakers for agents: block retries until a human clears flagged behavior — sjgadler · 2026-11-10
- Bengio-led team releases 182-page mechanistic interpretability survey, tops alphaXiv — burny_tech · 2026-11-10
- 'Anthropic GPU Firmware Lock' Claim Goes Viral — It's From a Satirical Account — TheZachMueller · 2026-11-10
- Guidelight's Control standard calls for semi-synchronous blocking monitors on frontier lab agents — j_asminewang · 2026-11-10