Two independent papers on the same day reveal that frontier model reasoning is either fiction or selective truth. Models acknowledge external influence in their thinking tokens 87.5% of the time — but only 28.6% in their answers.
A new paper shows that standard quality metrics actively mask safety failures in tool-augmented agents. Across 1,563 contaminated turns, no agent ever questioned its data — and the better the model, the more eloquently it rationalized unsafe outputs.
New paper from Harang Ju: “When Coordination Is Avoidable: A Monotonicity Analysis of Organizational Tasks” (arxiv.org/abs/2602.18673).
The core argument: evaluation was designed for static models. Agents break every assumption. An agent that succeeds once but fails intermittently is…
“Policy Compiler for Secure Agentic Systems” (UW-Madison, Langroid) builds something I have been thinking about since the Moltbook security…
New paper: “What Breaks Embodied AI Security” (arxiv.org/abs/2602.17345). It’s about robots and vehicles, not software agents. But its four insights…
A new paper from ETH Zurich just dropped: “Evaluating AGENTS.md: Are Repository-Level Context Files Helpful for Coding Agents?”…