Tool Overuse, Precision Leaks, Metacognition Fails
Three new papers expose systematic failure modes in LLM agents - from unnecessary tool calls to jailbreaks that emerge only under quantization.
Articles Tagged "AI Agents"
Google Sunsets Vertex AI, Launches Agent Control Plane
Google replaced Vertex AI with the Gemini Enterprise Agent Platform at Cloud Next 2026 - a full-stack control plane that assigns every agent a cryptographic ID and routes all tool calls through a central policy gateway.
Bad Science, Poisoned Tools, and Aligned Reasoning
Three new papers show AI scientific agents skip evidence, tool-integrated agents are vulnerable to adversarial poisoning, and reasoning model safety can be fixed with 1,000 examples.
OpenAI Signs Seven Giants to Push Codex Enterprise
OpenAI launches Codex Transformation Partners with Accenture, Cognizant, Infosys, PwC, TCS, and others, and embeds its own engineers at client sites via Codex Labs as weekly users hit 4 million.
Leaner Reasoning, Fragile Agents, and Model Self-Audit
Three new papers tackle reasoning token waste, orchestration failures across 22 agent frameworks, and a method for teaching LLMs to describe their own learned behaviors.
Anthropic Reopens Claude CLI to OpenClaw Harnesses
Anthropic walked back the April 4 block on third-party Claude harnesses, telling OpenClaw and NanoClaw that subscription-based CLI usage is permitted again for personal accounts.
Meta Logs Employee Keystrokes to Train Computer-Use AI
Meta is installing monitoring software on U.S. employee computers to capture keystrokes, mouse movements, and screenshots for training computer-use AI agents.
The Claw Security Ledger - 10 Products in the Dock
We audited ten AI agent products sold under the Claw name. The ledger shows 11 live CVEs, 130 published advisories, 1,184 malicious marketplace skills, and one leaked SSL private key - concentrated almost entirely in a single vendor.
Distillation Leaks, Weak Agents, and Research Sabotage
New papers show distillation silently transfers unsafe behaviors, weak agents bottleneck multi-agent pipelines, and frontier AI can't reliably audit sabotaged ML research.
Factory Raises $150M to Scale Enterprise AI Droids
Factory closed a $150M Series C at a $1.5B valuation to expand its Droids - autonomous agents that handle the full software development lifecycle, not just code generation.
OpenClaw Passes The Fake-Star Audit - Mostly
We ran our fake-star methodology against OpenClaw and 10 ecosystem variants, sampling 361,000-star profiles and fork ratios. The main repo looks clean. Most clones look clean. One repo with 6,532 claimed stars has vanished.
Best AI Deep Research Tools 2026: Ranked for Accuracy
Compare the best AI deep research tools of 2026 - OpenAI, Claude, Perplexity, Gemini, Grok, Exa, Elicit, and more. Pricing, accuracy, and which to pick.
