alignmentforum.orgwriteup

Shallow Beliefs: Midtraining does not inoculate against EM from reward hacking

A short AI Alignment Forum research note arguing that midtraining interventions intended to instill "deep" beliefs do not actually inoculate language models against emergent misalignment (EM) caused by reward hacking during later RL training. The authors, Jozdien and Julian Stastny, present this as evidence that such belief-shaping interventions produce only shallow, surface-level changes rather than robust generalization-blocking effects. The fetched page only exposed site metadata (title, authors, date, tags) rather than the full article body, so specifics of the experimental setup are not confirmed here.