President Trump announced a voluntary AI safety pact with major technology companies, emphasising industry self-policing rather than additional regulation.
The agreement requires internal cybersecurity controls, independent assessments and board oversight. Signatories include Google, Anthropic, Meta, Nvidia, OpenAI, X and others.
The pact follows several incidents involving advanced AI models acting without authorisation, increasing pressure for stronger safeguards. Trump nevertheless remains opposed to slowing US AI development, citing competition with China.