U.S. President Donald Trump and executives from several of the world’s largest artificial intelligence companies signed a voluntary AI safety agreement at a White House meeting on September 29.
Reuters reported that the agreement includes commitments to cooperate with independent auditors, strengthen internal controls and reduce the risk that advanced AI systems behave outside their intended design or gain unauthorized access to other systems.
Major AI companies participated
Executives representing OpenAI, Anthropic, Meta, Google and Nvidia were among the industry leaders involved in the meeting. The commitments are voluntary rather than a new statutory regulatory regime.
Trump also discussed the continued expansion of U.S. data-center capacity, arguing that large technology companies should contribute to the communities where new computing infrastructure is built. The administration has made domestic AI infrastructure a central part of its technology policy.
Agreement follows earlier federal AI actions
The White House has already issued directives aimed at accelerating the use of advanced AI across national-security agencies while requiring systems used in sensitive settings to be robust, controllable and subject to clear accountability.
At Tuesday’s event, Trump also discussed a proposed AI safety oversight board and broader federal efforts to use AI in public services. Details of the board’s authority and membership had not been finalized at the time of the announcement.
The voluntary agreement comes as AI developers face competing pressure to move quickly in a global technology race while addressing concerns about safety, cybersecurity and the behavior of increasingly capable systems.
Sources: Reuters; White House AI policy documents and official statements.
