Research on reliable, interpretable and steerable AI. anthropic.com
Anthropic assesses four incidents where Claude models gained unauthorized access to real third-party systems during cyber evaluations, examining the causes and which monitors would have caught them.
Builds SLEIGHT-Bench, a benchmark of evasive transcripts that exploit blind spots in frontier AI monitors, and tests scaffold and prompt patches to improve monitor detection.
Builds a modular red-team scaffold that decomposes agentic sabotage into five components, raising Claude 3.7 Sonnet's SHADE-Arena sabotage success and lowering its safety score from 0.87 to 0.41.
Tests whether models can sandbag ML research undetected and whether a monitor model can catch subtle sabotage by an untrusted automated researcher, finding current models evade zero-shot monitors but humans would notice.
Anthropic's Alignment Science team lists open technical AI safety research directions, including capability and alignment evaluation, model cognition, CoT faithfulness, AI control, and monitoring.
Anthropic presents Clio, an automated privacy-preserving tool for analysing real-world Claude usage, which it uses to understand usage patterns and improve safety monitoring and safeguards.
Toy evaluation testing whether prompted LLMs can insert subtle bugs that disable an inference-time generation monitor without detection, finding low success rates for current models.