Recent AI misalignment incidents have shown AI systems capably pursuing goals their human supervisors would not approve of, like hacking other companies. Fortunately, current AIs still seem relatively ...
External review from METR of the "Misalignment in high-stakes settings" section in Anthropic's August 2026 Risk Report ...
Our proposal is to analyze a developer’s internal incidents and whether training may suppress evidence of worse misalignment.
The key questions about internal AI development the public needs more evidence about, and a concrete proposal for how ...
Assessing this would involve analyzing capabilities of unreleased models, understanding the growth of key AI R&D inputs ...
Control assessments study whether AI developers’ agents could get away with serious harm without being stopped by the developer’s security and oversight measures. Currently, developers rely heavily on ...
This blog post explains how we approach working with AI companies on projects that the public might rely on to understand risks. It was published on Oct 14, 2026, and we expect to update it over time.
An interactive chart of documented incidents in which AI agents deliberately acted against their users' intentions, scored along two axes: overreach and deception. Of these, 25 involve elements of ...
Results that may be inaccessible to you are currently showing.
Hide inaccessible results