Research
16d agoApple Finds LLMs Struggle With Bayesian Logic
Researchers at Apple discovered that LLMs fail to update probabilistic beliefs consistently when presented with new evidence. They measured this via an "information processing gap" to quantify deviations from rational Bayesian updates. These internal inconsistencies hinder reliable decision-making in high-stakes fields.
The Signal
Practitioners should avoid relying on LLMs for precise probabilistic reasoning in medicine or law.