AI News – May 2 2026 1. Sam Altman drops support for universal basic income OpenAI CEO Sam Altman has shifted his stance on universal basic income. After funding multimillion-dollar experiments with his own money, he now believes fixed cash payments do not sufficiently address the deep labor market changes driven by advanced AI and is seeking more structural alternatives for the intelligence age. 2. Pentagon authorizes AI from 8 tech firms on classified networks The US Department of Defense signed agreements with eight leading companies to deploy advanced AI capabilities on its classified networks for operational use. OpenAI, xAI, Google and others are included, while Anthropic is excluded due to differences in safety policies. 3. Chinese court protects jobs from AI replacement A court in Hangzhou ruled that companies cannot legally fire workers solely to replace them with AI for cost savings. Adopting artificial intelligence is considered a voluntary business decision that does not justify dismissals without retraining or reasonable reassignment. 4. Neuralink automates neurosurgery with advanced robotics Neuralink is accelerating 2026 plans with high-volume brain implant production and nearly fully automated robotic surgery. The system improves precision, reduces time and increases scalability, enabling safer and more accessible implant procedures for more patients. 5. Google launches experimental AI assistant COSMO Google briefly released the experimental COSMO app for Android, powered by Gemini Nano for on-device processing even offline. It includes productivity tools for lists, documents and calendars. The app was quickly pulled, suggesting testing ahead of a larger launch.