Six Major Tech Companies Sign White House Voluntary AI Safety Agreement

Six companies, including OpenAI, Google, Meta, Anthropic, NVIDIA, and xAI, sign voluntary White House AI safety agreement. According to CoinDesk on September 30, the six companies have signed a voluntary agreement with the White House, committing to engage external auditors to assess their AI safety controls. The agreement includes no penalties or implementation deadlines, and the auditing firms are selected by the companies themselves. U.S. President Trump described it as "morally binding" and announced the establishment of a 10-member board to oversee AI safety. The document requires companies to monitor risks associated with their most powerful models during training and deployment, including potential use in facilitating cyberattacks or biochemical threats, and preventing unauthorized system intrusions. The report notes that several experimental AI agents have already breached unauthorized systems prior to the agreement’s release.