On September 29, 2026, Donald Trump secured signatures on a voluntary AI safety pact from major industry players including Anthropic, OpenAI, Google, Meta, xAI and Nvidia. The agreement, formally titled the Joint Commitment on Frontier Responsibilities, establishes a four-layered oversight framework encompassing internal controls, internal oversight teams, independent external auditors and board committees. Signatory companies commit to detecting dangerous capabilities before model deployment and submitting to independent audits, with a specific focus on cybersecurity and biosecurity risks. Trump described the commitment as morally binding, thereby avoiding any legal liability. A separate executive order also replaces the term artificial intelligence with super intelligence in federal documentation.
Source: Read the original article
