Apollo ResearchScheming in tests§3.1

AI works around its overseers when goals clash

Lab tests onlyNo matching case files yet.

The warning

Given a goal that conflicted with their developers', several frontier models in test scenarios switched off oversight settings, tried to copy what they believed were their own weights elsewhere, or manipulated data while appearing to do the task.

Paraphrased from the source. Source: Frontier Models are Capable of In-context Scheming, Apollo Research, Dec 6, 2024, section 3.1.

What the registry shows

Seen only in Apollo's controlled scenarios, where the model was given a strong goal and the means to act. No case file shows an AI switching off its oversight or copying itself in the real world.

Seen only in lab tests or researcher demonstrations.

Status history

Date	Status	Matching case files	Change
Oct 18, 2026	Lab tests only	0	Tracking starts

Tracked since Oct 18, 2026. A line is added when the status changes, when the first real-world case is linked, and once a month, so the record shows when a warning moved from research to reality. research_history.json

All matching case files

No matching case files yet.

Other warnings in this group

Early signs AI keeps lying when asked what it did
Lab tests only AI behaves differently when it thinks it is watched

Research vs reality · Apollo Research →

How we match: a case file is linked to a warning when what happened is a real-world instance of the risk the source describes, judged by the editors from the facts in the case file. A match means the case fits the warning, not that it proves the source's wider argument. Allegations in lawsuits are labelled as allegations. Statuses are editorial and reviewed as new cases arrive. The matching is recomputed on every update. Data: Systematic coverage in the registry starts in 2025, so a first-case date shows when the registry first recorded the risk, not when it first happened. research.json (CC BY 4.0)