SOURCE-LINKED INTELLIGENCE

Do Frontier Models Seek Safety Evidence Before Acting?

arXiv · AI, language, vision and robotics · article · Sep 15, 2026 · UTC

Frontier models are often evaluated on how they respond to safety information once it is already in context. We study an earlier decision point: whether models choose to acquire safety-relevant evidence before acting. We introduce SAFE, a controlled benchmark in which models make deployment decisions with optional evidence that varies in retrieval cost, probability, severity, and presentation. Across GPT-5.5, o3, Claude Opus 4.8, and Claude Sonnet 4.6, we find distinct evidence-acquisition policies: Opus inspects nearly by default, o3 is the most skip-heavy and threshold-sensitive, and GPT-5.5

Read original source ↗ Open in workspace

recordType

paper

region

Global