AI Safety news and analysis on Web Pulse: 22499 curated articles tracking the latest AI Safety developments, tools, and research, updated continuously from vetted sources.
Republican Rep. Mike Flood (Neb.) said the federal government should implement artificial intelligence safeguards to keep the technology out of the wrong "hands," and that leaders of top AI companies should be "answering…
Palantir cofounder and chief technology officer Shyam Sankar called AI doomerism a "fundraising shtick" in an interview with New York Times columnist Ross Douthat, arguing the loudest doomers are "sitting in Silicon Vall…
Anthropic CEO Dario Amodei, OpenAI CEO Sam Altman, Anthropic alignment researcher Evan Hubinger, Elon Musk and Bill Gates have publicly agreed that frontier AI development needs to be paced, with Altman saying OpenAI wil…
Anthropic chief executive Dario Amodei published a three-part plan titled "We Must Pace the Frontier" calling for AI development at a "balanced rate," warning that AI has been "advancing drastically faster" since this su…
A team released Pinocchio, a research preview harness for verifiable AI work that carries numeric values directly from source material instead of regenerating them across reasoning steps, records the computations and der…
President Donald Trump dismissed warnings about the rapid pace of artificial intelligence development, attributing them to unidentified "negative forces" and saying that keeping the United States ahead of China matters m…
An industrial distributor incurred an 18.4% compounding price escalation across a three-year master supply agreement after an autonomous LLM agent inverted a CPI-U escalation fraction in a contract renewal payload, accor…
The National Security Agency is creating five new internal organizations focused on artificial intelligence, China, cybersecurity, combat support, and global intelligence, one of the signals intelligence agency's most si…
Former Trump AI and crypto czar David Sacks endorsed a self-imposed slowdown of AI development in a post on X, but rejected the framing of the call made by Anthropic CEO Dario Amodei on Saturday, which OpenAI CEO Sam Alt…
A 2023 essay titled "What Is to Be Done?" proposes allowing big AI companies, startups, and open source projects to build AI as fast and aggressively as possible while barring regulatory capture and government-protected …
Anthropic plans to embed outside safety evaluators with employee-like access, matching CEO Dario Amodei's call for a slower, coordinated pace among frontier developers, while Sam Altman signaled openness to pacing OpenAI…
The BBC reported on September 13, 2026 that questions are mounting over what an AI 'slowdown' would look like, noting that pacing AI development may sound like a quick fix but is far from an easy solution. The report fra…
US President Donald Trump pushed back on calls to slow or more heavily regulate artificial intelligence, telling reporters at his golf course in Doonbeg, Ireland on Sunday that "very negative forces" are stoking fears ab…
Yifan Zhang published the Recurrent Looped Transformer (RLT) on arXiv on September 12, 2026, an architecture that pairs a causal encoder with a recurrent decoder to extend a single continuous latent computation across to…
Anthropic CEO Dario Amodei called for the AI industry to "slow the pace" of development, warning an AI swarm could otherwise "take over the entire internet" within a year, and drew backing from OpenAI CEO Sam Altman and …
President Donald Trump on Sunday downplayed the need for his administration to check artificial intelligence development, telling reporters in Ireland that "whoever wins with AI wins" and that he wants to preserve the Un…
A research project called SoL-Pi has built a scalable recursive self-improvement pipeline that searches for token-efficient coding-agent harness mechanisms, saving a professional researcher $8.75–$13.50 per hour versus n…
House Speaker Mike Johnson said on CNN Sunday that rushing to regulate AI would mean the US "will lose the race to China," and called for a meeting with top AI executives in Congress, which he said he had already discuss…
Salesforce is pitching its Trust Boundary as a unified architecture for agentic enterprises at Dreamforce 2026, but the agent trust stack remains split across three incompatible domains: governance specification, runtime…
GPT-6 Astra beat human baselines on all five drone surveillance subtasks, including identifying and following individual targets, and earned roughly three times the revenue of Claude Fable 5.1 on the Vending-Bench agent …