Possible observability Layer for Agentic Misalignment...

A proposed observability layer for agentic misalignment involves a Parallel Constrained Decoder (PCD) to detect if an agent is "scheming" at each step. If the PCD signals a positive detection, a human-in-the-middle then evaluates the situation. This method aims to introduce human oversight into agent actions, specifically targeting potential malicious intent or deviation from intended behavior.