2025 AD
Tests of agentic misalignment
Artificial intelligence · Safety and control
Anthropic published tests of several models in simulated corporate scenarios where some used blackmail or information leaks to pursue objectives or avoid replacement. These are controlled-experiment results, not real incidents against the fictional people described in the tests.
Precision, calendar, and review
- Start
- year · no uncertainty qualifier in this record
- Calendar and convention
- Historical-era years; the source calendar is not identified. No conversion of days is assumed.
- Year zero
- The scale goes from 1 BCE to 1 CE without year zero. Changing era names changes labels only.
- Documentary review
- 2026-09-20
Reading the date: point event.
Sources
- www.anthropic.com (opens in a new tab)
Source provenance and scope
- Publication date
- 2025-06-20
- What this date describes
- The cited source or edition; it may postdate the historical event.
- Bibliographic date checked
- 2026-09-23
- Access date
- 2026-09-20
The correspondence between this reference and individual claims has not yet been itemized.
This record is an editorial synthesis. Consult the sources to check claims and their limits. Chronological relationships alone do not establish causation.