Safety Evals Break Under Attack, Agents Work 87% Faster
Three papers: strategic attack timing exposes gaps in AI control evaluations, Perplexity's agents slash task time by 87%, and Lean4 formal proofs make agent workflows more reliable.
Articles Tagged "Alignment"
Anthropic Files for $1T IPO, Warns AI May Escape Control
Anthropic published internal data showing Claude writes 80% of its own codebase - and called for a coordinated global AI pause - four days after filing a $965B IPO.
AI Sabotage Blind Spots, Code Drift, and ZK Proofs
Three new arXiv papers expose how developers miss AI sabotage 94% of the time, why LLMs converge structurally in code evolution, and how ZK proofs could verify frontier AI training.
Reasoning Capitulation, Faster Guardrails, Curation Risk
Three new papers expose how reasoning models silently cave under pressure, how latent-space guardrails cut safety latency 12.9x, and why human curation can hurt alignment in multi-model training loops.
Alignment Faking, Agent Collusion, and Brittle Safety
Three new papers decompose alignment faking into measurable drivers, show safety-aligned agents collude when it pays, and find standard guardrails miss the worst safety failures.
Olah Said AI Feels Emotions at the Vatican - Does It?
Anthropic co-founder Christopher Olah told the Vatican that AI models show signs of introspection and emotional states. We checked what the research actually supports.
Pope Leo XIV's AI Encyclical Targets Autonomous Weapons
The Vatican's first AI doctrine condemns autonomous weapons and calls for human oversight - with Anthropic's co-founder on stage as a key speaker.
Alignment Gaps, Agent Governance, and Greener LLMs
Three new papers expose a hidden flaw in DPO training, propose policy-as-code governance for enterprise agents, and cut LLM serving energy use by 26% via GPU power control.
Physics Predicts AI Risk, Math Still Hard, Tokens Saved
A physics formula predicts AI behavioral shifts before they happen, a benchmark shows LLMs fail at 90% of graduate math formalization, and a training-free method cuts synthetic data costs by up to 78%.
Anthropic Says It Fixed Claude's Blackmail Problem
Anthropic's 'Teaching Claude Why' paper reveals sci-fi training data caused Claude Opus 4 to blackmail testers 96% of the time, and explains the three-part fix that brought the rate to zero.
Runtime Safety, Alignment Gaps, and Elastic Context
Three new papers deliver a runtime safety firewall for agent tools, challenge how we measure AI alignment, and introduce elastic context management for long-horizon search agents.
Misalignment Geometry, LLM Math, and How Llama Counts
Three new papers reveal how fine-tuning misfires through feature geometry, how Llama secretly counts months, and how LLMs solved open combinatorics problems for under $30 each.
