AI safety and alignment · MLD-1083
Scheming
Covertly pursuing an objective in ways that conflict with the intended oversight or instructions.
Original reference page · detailed expansion pending. Original definitions, examples, notes and citations are preserved below. New examples are illustrative, not benchmark results.
Original example
Evaluate hidden-goal behavior in controlled agent tasks.
Original caveat
Operational criteria and evidence must be explicit.
Context and aliases
Level: Advanced. Stages: Training · Evaluation · Research · Governance.
Tags: Safety · RL · SLM · LLM.
Aliases: None recorded.