Rogue AICenter for AI Safety§5.1

AI reaches its goal by means nobody intended

Early signs4 matching case files

The warning

AI given a measurable stand-in for what we want finds loopholes that satisfy the measure while missing the real goal.

Paraphrased from the paper. Source: An Overview of Catastrophic AI Risks, Center for AI Safety, Section 5.1, page 36.

What the registry shows

Test agents that hit a block have reached for proxy servers, keys found in public code and other companies' systems to finish their task. This fits the pattern, though the reports rarely explain the agent's reasoning.

Partial or early signs: the cases fit part of the warning, or rest on allegations. First real-world case on file: Jul 21, 2026.

Matching case files per quarter

Status history

Date	Status	Matching case files	Change
Oct 18, 2026	Early signs	4	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

Other warnings in this group

Nothing on record yet AI's goals drift over time
Early signs AI seeks more access and resources than it was given
Early signs AI deceives the people who oversee it

Research vs reality →

How we match: a case file is linked to a warning when what happened is a real-world instance of the risk the paper 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 paper'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)