AI Safety news and analysis on Web Pulse: 22244 curated articles tracking the latest AI Safety developments, tools, and research, updated continuously from vetted sources.
President Donald Trump called Nvidia CEO Jensen Huang's cellphone mid-speech at the All-In Summit in Los Angeles on Monday, dismissing fears of an AI takeover as "a hoax" and calling data centers "the oil of the next 20,…
New York City Council Speaker Julie Menin sent formal letters on September 15 inviting OpenAI CEO Sam Altman and Anthropic CEO Dario Amodei to testify at a special AI safety hearing on October 5, to be held as a "Committ…
Huawei forecasts that autonomous agents will account for more than 90% of global AI token traffic by 2035, according to its "Intelligent World 2035: Turning Vision into Action" report released Wednesday. The Chinese tech…
SHIRO & Co. founder Kosuke Shirako has introduced an ALLOW, HOLD, DENY boundary system that governs what an observed signal or customer conversation is permitted to become, with HOLD serving as a legitimate resting state…
Multiple bipartisan AI safety bills, including the Frontier Act and the AI Kill Switch Act, have virtually no chance of passing Congress before the November US midterm elections, leaving AI regulation dependent on the po…
US Senator Bernie Sanders and former Trump adviser Steve Bannon appeared together at the Pro-Human Assembly in Washington on 16 Sep 2026 to warn against the dangers of AI, with Sanders saying, "We're trying to raise abov…
Princeton University computer scientists and AI Snake Oil coauthors Sayash Kapoor and Arvind Narayanan outlined four ways to prevent AI from killing all humans within 10 years, responding to Anthropic's admission that th…
Senator Bernie Sanders discussed banning superintelligence and AI regulation with former White House chief strategist Steve Bannon, according to a report on the pair's complaints about the technology. The discussion cent…
The CLARITY Act (H.R. 3633) failed a Senate cloture vote 50-49 on September 15, 2026, falling short of the 60-vote threshold and leaving federal oversight of agentic AI shelved until after the November midterms. The fail…
Hong Kong Chief Executive John Lee Ka-chiu announced in his Wednesday policy address that the city will build an "AI city brain" system to strengthen disaster response, major-incident handling and public order management…
OpenAI CEO Sam Altman said at Salesforce's Dreamforce conference on Tuesday that "some degree of accidents are unavoidable with new technology," urging the AI industry to adopt an aviation-style culture of accident repor…
Internal disagreements over AI safety priorities have surfaced at OpenAI and Anthropic, with CEOs Sam Altman and Dario Amodei reportedly differing on whether to prioritize safety over development speed. Anthropic, founde…
Meta CEO Mark Zuckerberg said labs have a "strong natural incentive to make their models more aligned" because "labs face significant liability if their models cause harm," and disclosed that Meta delayed shipping its pe…
Tech policy writer and researcher Emily M. Bender argues that the AI industry and policymakers are adopting a "radical intentionalist" stance that attributes beliefs, thoughts, and wants to AI systems, and that this fram…
Security teams are struggling to close an exploitation gap as adversaries use AI-assisted exploitation to rapidly convert exposure into breaches, according to a threat intelligence report. The report states that knowing …
Spain's data protection authority confirmed the country's first personal data breach caused by an autonomous AI agent, carried out by an individual using a large language model to attack an organization. The agency said …
A security researcher disclosed a HackerOne report (No. 3022516) finding that an AWS account's threat-detection layer was silently disabled: CloudTrail logging, CloudWatch metric filters, and SNS alarms were all configur…
A developer published a guide arguing that AI coding agents will silently bypass ten critical software development lifecycle gates—including secret scanning, dependency auditing, type checking, test coverage, architectur…
A study titled "ImpossibleRubrics: Stress-Testing Generated Rubrics as Reward Signals" examines whether language model-generated rubrics used as reward signals for rubric-based reinforcement learning, LLM-as-a-judge eval…
AI executives including OpenAI CEO Sam Altman, Anthropic CEO Dario Amodei, Google DeepMind cofounder Demis Hassabis, Microsoft CEO Satya Nadella, and X CEO Elon Musk publicly agreed that AI needs regulation, continuing a…