Where AI Agents Break: Research, Safety, and Privacy
Three new papers expose where autonomous agents still fail: fabricating research, turning hallucinations into security exploits, and leaking private data from small models.
Science Articles
Fix 8% of Tokens, Dodge Memory Attacks, Cut Agent Costs
New research pinpoints the 8% of tokens driving reasoning failures, exposes memory laundering in agent systems, and cuts web agent inference costs 1.9x.
Self-Correcting Models, Smarter Monitors, AI Designs Itself
Three new papers tackle critique dependency in LLMs, ensemble monitoring for AI control, and agents that autonomously discover better neural architectures.
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%.
Olympiad Gold, Broken Memories, and Attention Loss
A 30B model earns IMO gold, memory consolidation silently corrupts agents, and a new metric predicts when LLMs lose track of their instructions.
Reasoning Bias, Behavior Cues, and Tool Interpretability
New research shows reasoning length amplifies position bias, behavior cues cut wasted tokens by 50% while boosting safety, and sparse autoencoders can predict tool failures from model internals.
Agent Overload, Blind Attention, Unsafe Traces
Three new papers show that more agent components backfire, reasoning models hide unsafe thinking, and vision-language models waste most of their attention.
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.
Agent Memory in 2026: Circuits, Tiers, Evolution
Three new papers reveal how agent memory silently breaks, how a tiered architecture recovers it, and how models can self-improve without human labels.
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.
Tool-Use Tax, Jailbreak Risk, and Robot Vision
Three new papers: tools slow LLM agents under noisy prompts, jailbreaks barely dent frontier model capabilities, and interleaved text-vision traces push robot success to 95.5%.
Prompt Traps, Swarm Failures, and AI-Discovered Physics
Three new papers reveal when few-shot examples hurt scientific reasoning, why homogeneous agent swarms lock in errors, and how an AI autonomously found a novel physical mechanism.
