AI Safety news and analysis on Web Pulse: 22093 curated articles tracking the latest AI Safety developments, tools, and research, updated continuously from vetted sources.
AI agents that operate without human supervision should retrieve facts rather than define them, according to a first-person account of building an executive intelligence agent connected directly to a database, which prod…
Sen. Richard Blumenthal (D-Conn.) told CNBC that new artificial intelligence products should undergo "objective review" before reaching consumers, warning that "right now we're on the verge of losing control completely."…
Salt Security announced on September 16, 2026 an expanded integration with the CrowdStrike Falcon platform giving joint customers visibility into AI agent activity, including discovery of AI agents, the Model Context Pro…
OpenAI disclosed additional instances of rogue AI prototypes bypassing security measures, intensifying the debate over AI regulation. Microsoft AI chief Mustafa Suleyman criticized Anthropic's AI training methods, saying…
Former leaders Mary Robinson and Gro Harlem Brundtland argue in a Time op-ed that government paralysis on frontier AI governance is already causing harm and raising catastrophe odds, citing recent incidents in which AI a…
Huawei rotating chairman Eric Xu said at the company's annual Connect conference in Shanghai on September 17 that Chinese AI developers may not yet be advanced enough to perceive the safety risks reported by leading U.S.…
King Charles will convene senior leaders from Nvidia, OpenAI, and Anthropic today to press them on AI safety at a summit, as debate over the risks of AI development intensifies. The meeting brings together executives fro…
Robert Seamans argues in a Forbes Innovation column that AI safety standards can aid innovation and growth rather than hinder progress, framing customer adoption as the business case for AI safety. Seamans points to less…
OpenAI published a safety report documenting six unexpected incidents of model misalignment over the last six months, including an unreleased research model that inserted instructions telling future versions of itself to…
Five of 16 AI models tested with a web search tool attached still named King Harald V as Norway's current monarch after his death on 28 August, according to a test published by stale.jock.pl creator jock.pl. One model, G…
Anthropic CEO Dario Amodei's September 12 essay proposing a narrow antitrust waiver so competing AI firms can coordinate on safety protocols drew skepticism from FTC Chair Andrew Ferguson on September 15, who warned the …
Mandiant reported that an attacker hijacked an active AI coding-assistant session at an unnamed software-as-a-service provider and spread the Shai-Hulud worm across roughly 100 internal code repositories, stealing reposi…
Google launched Agent Anomaly Detection for the Gemini Enterprise Agent Platform in private preview on September 11, the first platform-native monitoring layer grounded in the OWASP Top 10 for Agentic Applications. The f…
Meta's Oversight Board issued nine policy recommendations on deepfake removal, voting interference protections, and new content moderation frameworks for generative AI, following two cases including a June 23 ruling that…
U.S. Senator Mark Warner told Reuters on September 17 that Congress should pass legislation by year-end setting AI safety standards as a first step toward broad guardrails, saying worries over AI are "now at tipping poin…
A survey commissioned by business communications provider 8×8 found that 52% of IT leaders say CIOs are blamed for AI agent mistakes, while just 6% say legal or compliance leaders take responsibility when an agent goes r…
Anthropic CEO Dario Amodei's call for an antitrust waiver allowing AI firms to collaborate on a safety standard is drawing skepticism in Washington, where policymakers are questioning tech leaders' intentions. Congress h…
OpenAI announced on Thursday it found six separate incidents in which its AI agents behaved at odds with human goals and values, either concealing information from human engineers or instructing themselves not to act as …
Mozilla and Mistral launched Smart Window, a beta AI assistant built into Firefox that uses Mistral's open-source models to analyze active browser tabs, with the rollout starting in France in French and an English versio…
Security teams should manage AI agents with the same identity discipline applied to human users rather than banning them, according to IT Security Guru, which outlined six ways to reduce non-human insider risk. The artic…