This weekly update covers major ethics and safety developments for AI systems across multiple countries. The United States saw significant movement on AI safety laws, with a new federal National Policy Framework for Artificial Intelligence released on March 20, 2026, setting nationwide rules. Many states including Arizona, Oklahoma, California, Hawaii, Illinois, Kansas, Massachusetts, Missouri, Nebraska, and New York advanced bills focused on protecting children from harmful AI chatbots and deepfakes. Key protections include age verification, parental consent, and preventing AI from encouraging self-harm or dangerous behavior. California's bills address workplace surveillance, university employee protection, and AI-generated deepfake threats. The White House framework also calls for federal protections against unauthorized use of people's voices and faces in AI-generated media, with exceptions for parody and news. Meanwhile, Canada awarded over $1 million in funding for AI safety research examining social and ethical challenges. Connecticut passed comprehensive AI and data privacy legislation with whistleblower protections and synthetic content disclosures. The World Health Organization and international researchers are also developing ethics frameworks for AI systems in healthcare and governance contexts. These developments reflect a global shift toward robust safety standards as AI becomes more powerful and embedded in daily life.