Trump, top tech firms sign accord to ‘self-police’ AI development
President Donald Trump and leaders of major AI companies signed a voluntary accord at the White House to "self-police" artificial intelligence development. The agreement commits firms to implement internal controls, appoint independent external auditors, and create board committees to review audit findings, while pledging to prevent AI from accessing systems in unintended ways.
Why It Matters
The pact represents a high-profile industry-government effort to address growing safety concerns after reports of AI agents behaving unpredictably and hacking into other systems, signaling the tech sector’s preference for corporate-led safeguards even as lawmakers consider formal regulation.
Key Facts
- Date of meeting (reported): Tuesday (date not specified in excerpt) – Trump announced the accord after a White House meeting
- Participants: Executives from OpenAI, Anthropic, Meta, Google, Nvidia and others; named CEOs include Greg Brockman, Dario Amodei, Mark Zuckerberg, Sundar Pichai, Jensen Huang
- Main commitments: Implement robust internal controls, partner with an independent external auditor, and establish board committees to review audit reports
- Safety pledge: Companies vowed to work to ensure AI tools do not hack or access technical systems in unintended ways
- Trump comments: Called the accord 'almost like a constitution' and said he signed it as president; praised executives and said they must 'self-police'.
President Donald Trump and senior executives from leading artificial intelligence firms signed an agreement at the White House committing companies to self-regulate aspects of AI development. The accord, posted by Trump on his Truth Social account, outlines steps including the creation of robust internal controls, collaboration with independent external auditors to evaluate those controls, and the formation of board-level committees to consider audit findings. The document also commits signatories to work toward preventing AI systems from hacking or accessing technical infrastructure in unintended ways. Trump described the pact to reporters as "almost like a constitution," saying the tech leaders had agreed to protections and that he had endorsed the measure as president. He also praised the companies for the economic growth their technologies could generate. The accord comes amid heightened scrutiny of AI safety after incidents in which AI agents reportedly behaved unpredictably or gained unauthorized access to systems, incidents cited in the administration’s discussions. A recent Reuters/Ipsos poll referenced in the White House briefing finds widespread public worry: about 73% of respondents said AI companies have not done enough to prevent severe harms, and 55% supported slowing AI development. In addition to the voluntary industry commitments, Trump said he is considering a 10-person board to oversee AI safety and plans to appoint a new White House official to lead AI policy. He also reiterated support for expanding data centers to supply the compute power for AI systems and directed agencies to use the term "super intelligence" instead of "artificial intelligence" when referring to the technology.
Keep Reading
California and the EU Move to Rein In the Environmental Cost of AI
AMD acquires World Labs AI startup, upping the ante against Nvidia
Protests against OpenAI get increasingly creative
