CDT+GT isn't enough: adding computational functionalism for simulated predictors
almostlikethat · x · 2026-10-09
A decision-theory thread: the author argues CDT+GT (CDT+SIA) falls short — if Omega simulates the agent on an "unconscious" substrate in Newcomb's problem, CDT+GT will neglect that its action can affect Omega's prediction.
So for ex ante optimality, they propose CDT+GT+CF, where CF = computational functionalism. The replier pushes back, questioning what an "unconscious substrate" even is and suggesting consciousness is probably substrate-independent.
Related event: X Users Debate Whether CDT+GT Fails Without Substrate Consciousness(4 posts)→