132 articles
AI safety, alignment, and governance — model risk, red-teaming, regulation, and policy. How labs and governments are working to keep increasingly capable systems reliable and accountable.
The Trump administration has engaged in multiple calls with Anthropic, led by cofounder Tom Brown, after replacing CEO Dario Amodei in discussions about re-releasing the Claude Fable 5 AI model.
Pangram CEO Max Spero claims language models expose themselves by repeating similar arguments, a key flaw in AI-generated text detection.
OpenAI announced Patch the Planet, a project offering free security consulting to open-source projects, with over 30 projects already participating.
The US banned foreign access to Anthropic's latest models, citing concerns over AI risks, with critics blaming the company's frequent warnings about potential harms.
The Trump administration ordered Anthropic to shut down its Fable 5 and Mythos 5 models, citing national security concerns. The order, issued on June 17, 2026, prompted immediate action from the company.
Signal President Meredith Whittaker warned that AI chatbots like ChatGPT and Claude are not your friends, citing privacy risks in a June 20, 2026 interview.
Eurocommerce, representing major retailers, seeks to exclude AI-generated ads from EU AI Act transparency rules, citing non-deceptive intent.
The White House ordered Anthropic to restrict export of its AI models Fable and Mythos, citing national security concerns, after the company pulled them from global access within a week.
The US government banned Anthropic's Fable 5 and Mythos 5 models due to national security concerns, sparking debate over whether the move could inadvertently benefit the company.
Norway will ban generative AI tools in elementary schools starting late August 2026, citing concerns over learning outcomes.
Google plans to appeal a Munich court ruling that held it directly liable for AI-generated search overviews, citing a Berlin court's opposite conclusion from June 2026.
The U.S. government has forced Anthropic to pull its advanced AI models offline due to export control concerns, citing risks of foreign access.