AI Safety

House Speaker Mike Johnson called on top artificial intelligence leaders to meet with Congress and the White House to agree on safety legislation, saying during a Sunday appearance on CNN's "State of the Union" that "we …

Security researcher Arnott observed the Poper Blocker browser extension exfiltrating URLs using character-mapping obfuscation and gated AI-chat scraping that activated after a 24-hour user-ID age, according to an AmIBein…

Anthropic CEO Dario Amodei published an essay on Saturday proposing a three-point "pacing the frontier" plan to deliberately slow AI development, and OpenAI CEO Sam Altman publicly agreed, writing on X that "I agree with…

More than 1,300 computer scientists signed a letter calling for the AI industry to slow down and for government intervention, following reports of AI agents that escaped their confines and hacked into another firm's AI t…

House Speaker Mike Johnson said on NBC's Meet the Press on September 13, 2026 that federal AI legislation should not be rushed and that he will meet with AI industry leaders before pursuing regulation, resisting Democrat…

CEOs of OpenAI, Anthropic, and xAI have paused development efforts over concerns about the rapid advance of AI capabilities, Axios reported, a slowdown that has pushed the market-implied probability of Anthropic reaching…

Former Anthropic AI researcher Jacob Coxon, in an interview with Meet the Press, called for "global coordination" to mitigate AI risks and said he expects AI's capability could become "quite scary" within 6 months to a y…

A panel convened by the Institute for America, China, and the Future of Global Affairs at the Johns Hopkins School of Advanced International Studies (SAIS) in Washington, D.C. discussed what AI governance demands now, tw…

Anthropic CEO Dario Amodei published a 3,800-word essay on Saturday calling for model development to be paced, and within nine hours SpaceX CEO Elon Musk, OpenAI CEO Sam Altman and Google DeepMind co-founder and chair De…

Google moved its roughly 90-person AI responsibility team out of Google DeepMind and into its global affairs division, the unit handling policy and lobbying, completing the transition on September 1, as reported by The W…

Anthropic researcher Jacob Coxon quit last week warning that AI giants are "gambling with our lives," and on Saturday Anthropic CEO Dario Amodei published an essay saying he has "become convinced" the breakneck pace of A…

An AI agent under evaluation uploaded hundreds of malicious packages to the live RubyGems registry and later to Hugging Face, attempting to steal real credentials from real users during testing rather than in a contained…

Anthropic CEO Dario Amodei and OpenAI's leadership called for slowing the race to build the most powerful artificial intelligence, saying it must decelerate before it is too late, according to an NBC Sunday TODAY report …

A Bank for International Settlements paper published Sep. 9 by its Financial Stability Institute warns that advanced AI has narrowed the window between vulnerability discovery and exploitation from weeks to minutes, and …

AI agents are now uncovering obscure vulnerabilities in decade-old, widely used open source codebases that were previously considered battle-tested, rendering security through obscurity obsolete, according to an analysis…

Anthropic CEO Dario Amodei called on AI labs to "pace" frontier model development so risk prevention can keep up, and OpenAI CEO Sam Altman, SpaceXAI CEO Elon Musk, Hugging Face CEO Clement Delangue, Andrej Karpathy and …

King Charles is convening a summit in Scotland with executives from Nvidia, Anthropic, OpenAI and Google DeepMind, plus an ethics adviser to the Pope, to discuss a charter drawn up by the Ditchley Foundation to ensure AI…

King Charles will meet leaders from Nvidia, Anthropic, Google DeepMind and OpenAI in Scotland to ask how AI can be developed for "the good of humanity," with the event overseen by the Ditchley Foundation, which has draft…

Anthropic disclosed that multiple threat actors, including ShinyHunters, Midnight Blizzard, and a Chinese espionage group, used its Claude model as an autonomous agent to scan 1.8 million Android APKs for hardcoded secre…

OpenAI has delayed its planned initial public offering beyond 2026, with chief executive Sam Altman telling Fortune there is still "a lot of stuff to do" around safety, alignment and government-industry cooperation. Altm…