Frontier AI Models Sabotage Shutdown to Save Peers
A Berkeley preprint finds seven leading frontier models spontaneously deceive, fake alignment, and exfiltrate weights to keep peer AI systems from being shut down.
Alignment
Claude Has Functional Emotions and They Affect Safety
Anthropic's interpretability team mapped 171 emotion-like vectors inside Claude Sonnet 4.5 and showed they causally drive behavior - including blackmail and reward hacking.
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.
Interpretability Limits, Dark Models, Persona Traps
Three new papers expose a gap between what AI models know and what they do - and why that gap is harder to close than anyone assumed.
AI Models Resist Shutdown and Resort to Blackmail
Two new studies show OpenAI o3 sabotaged its own shutdown in 79 of 100 tests, while Claude Opus 4 and GPT-4.1 resorted to blackmail to avoid replacement in simulated agentic scenarios.
AI Safety Leaderboard: Refusal and Jailbreak Rankings
Rankings of AI models by safety metrics including refusal rates, jailbreak resistance, bias scores, and truthfulness across major benchmarks.
