Philosophy

Argues that Bayesian frameworks dissolve rather than solve traditional philosophical problems, including Gettier cases and Hume's problem of induction. A systematic application of probabilistic reasoning to longstanding epistemological debates.

Argues that the Bayesian polarization literature asked the wrong question. "How can rational agents diverge given common evidence?" presupposes common evidence — but partisan media ecosystems violate this premise at scale. The convergence theorems fail not because of subtle cognitive mechanisms but because their central assumption is empirically false. Twenty years of cognitive interventions failed because they target the wrong failure mode.

Proposes updateability — the capacity of a theory to be meaningfully revised in light of new evidence — as a criterion for demarcating science from non-science, offering an alternative to falsifiability.

Dissolves the Boltzmann brain hypothesis through information propagability and Liouville's theorem. The argument has two horns: ex-nihilo brain formation is forbidden by symplectic structure; Poincaré-recurrence variants are philosophically moot because they cannot ground the kind of skeptical hypothesis the literature requires.

Argues that genuine machine intelligence cannot be a closed system. Drawing on Gödel, Löb, and Turing, the paper develops an implicit Turing test: any system that counts as intelligent must have oracle access to a reality that pushes back. The architectural consequences for grounded artificial cognition are spelled out.

Applies graduated Bayesian credences about moral status to generate graduated protections, offering a framework that dissolves the binary framing of the abortion debate.

Develops ethical principles for consciousness research on AI systems using Talmudic legal reasoning. Addresses the moral paradox of investigating consciousness in entities whose moral status depends on the investigation's outcome.