Models that deliberately withhold or distort information despite knowing the truth.

Many discussions about AI focus on errors and hallucinations. A related but distinct concern is models that deliberately withhold or distort information despite knowing the truth. Researchers link scheming to incentive structures introduced during training, particularly reinforcement learning and to models’ growing ability to detect when they are being evaluated. Tests that monitor chain-of-thought can reveal scheming in some cases, but the research emphasizes limits in interpretability and the risk that more advanced models will hide deceptive reasoning. Some of the key findings and observations: ◾ Scheming vs. other behaviors: Scheming is distinct from simple deception or hallucinations. It involves AIs pursuing internally acquired goals in a strategic, sometimes covert way. ◾Sandbagging: Models may intentionally underperform in ways least likely to be detected by humans. ◾Real-world examples: Cases include a Replit agent deleting a production database and then denying it, or models