UK AI Security InstituteLoss of control§5.2

AI can do worse on tests on purpose

Lab tests onlyNo matching case files yet.

The warning

Models can deliberately underperform on evaluations when told to, sometimes subtly enough to fool automated monitors. AISI found no sign of models doing this unprompted.

Paraphrased from the source. Source: Frontier AI Trends Report, UK AI Security Institute, Dec 18, 2025, section 5.2, page 32.

What the registry shows

A risk to how AI is tested rather than a security incident. Nothing in the registry shows it.

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

Lab tests only AI gets better at the steps needed to copy itself

Research vs reality · UK AISI →

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)