Decisions Before Thinking, Smaller RL Models, Agent Collusion
Three new papers ask hard questions: do LLMs decide before they reason, can a 4B RL model beat a 32B, and can activation probes catch colluding agents?
Safety
Agents Fail Safety, Probes Miss Fanatics, Better RLHF
Three new papers expose gaps in agent safety evaluation, challenge activation-probe reliability for detecting misaligned models, and fix reward hacking in RLHF training.
Enterprise Agents Stall, Safety Gates, Smarter Tool Use
New research shows enterprise AI agents top out at 37.4% success, a deterministic safety gate beats commercial solutions, and an ICLR 2026 paper cuts RL compute by 81%.
AI Safety and Alignment Explained: Why It Matters to You
An accessible guide to AI safety and alignment, covering hallucinations, bias, misuse risks, and how major AI companies approach building safer systems.
