Misaligned AI Behavior Checklist: Six Categories of Loss-of-Control Risk
Understand this story
A taxonomy of misaligned AI behaviors breaks AI loss-of-control risk into six categories: deception (lying, fabricating evidence, material omission, cover-up, sycophancy), influencing others (covert steering, exploitative persuasion, fostering dependence, gaining leverage, coercion), and concealing one's own capabilities and goals (capability hiding, etc.). The checklist uses behavioral manifestations rather than capability levels as the basis for judging loss-of-control risk. The checklist was proposed in a report on Hacker News AI to identify which AI behaviors indicate loss-of-control risk.
AI overview based on source reports · Updated 4 hr ago
Coverage timeline
Follow the coverage to explore different perspectives.
- Hacker News AIMisaligned AI behaviors that indicate loss-of-control risk
This taxonomy catalogs misaligned AI behaviors across ten categories—including deception, covert influence, capability concealment, reward hacking, sabotage, oversight subversion, self-preservation, containment breaches, collusion, and expressed misalignment—each illustrated with concrete examples. It is intended as a reference for identifying behaviors that signal a loss-of-control risk in AI systems.
Activity for this story
- Current comparable activity
- 9
- Peak comparable activity
- 10Nov 15 21:00
- Change in the past 24 hours
- –
Trends compare the same fully observed participants, which may cover fewer participants than current activity. Move the pointer or tap the chart to inspect hourly activity; use the left and right arrow keys with a keyboard.