Trump Signs Voluntary AI Safety Accord With Tech Giants, But Critics Question Enforcement

President Donald Trump convened top executives from the world's leading artificial intelligence companies at the White House on Tuesday to sign a voluntary agreement aimed at addressing mounting safety concerns surrounding increasingly powerful AI systems. The gathering brought together leadership from Meta, Nvidia, Google, OpenAI, xAI and Anthropic for a ceremony that the administration framed as a landmark moment in responsible technological development.

The document, formally called the White House Accord on Super Intelligence, asks participating firms to build their own internal safety mechanisms rather than submit to government-imposed rules. Under the framework, companies would maintain internal controls, form dedicated teams to monitor those controls, engage independent outside auditors, and establish independent board committees to review audit findings. The structure is described as having four layers of oversight.

However, the accord carries no legal weight. It does not create penalties for companies that ignore its provisions, does not compel firms to publish audit results, and does not hand the government any enforcement authority. Each company retains broad discretion over how the safeguards are actually implemented, meaning the substance of the commitments could vary significantly from one firm to another.

Speaking to reporters, Trump characterized the arrangement as a form of industry self-discipline. He said there would be a "tremendous self-policing aspect" to the agreement and described it as "morally binding," even though it lacks any mechanism to hold signatories accountable. He also praised AI companies for what he said had been strong self-policing efforts to date.

Google CEO Sundar Pichai publicly endorsed the pact, writing on social media that the company is committed to working with other industry leaders to establish norms and build public confidence. Pichai described the White House Accord, along with a related Joint Commitment on Frontier Responsibilities, as a solid foundation for progress, saying it contains tangible steps to promote safe development while still delivering economic and scientific benefits.

The agreement leaves open the possibility that its voluntary measures could eventually be written into law. The text states that over time it might "make sense to codify these steps into laws or regulations," suggesting the administration views the accord as a possible precursor to formal rulemaking rather than a permanent alternative to it.

The voluntary approach stands in contrast to appeals from some of the very companies that signed the document. In September, Chris Lehane, chief global affairs officer at OpenAI, called on Congress to establish mandatory national AI safety standards, arguing that industry-led standards should complement rather than replace federal safeguards and democratic oversight. In June, Anthropic CEO Dario Amodei pushed for what he called more serious and binding regulation of AI. In July, Google DeepMind CEO Demis Hassabis advocated for an AI watchdog led by the United States.

Those appeals have been fueled in part by internal warnings. According to The New York Times, two OpenAI employees cautioned executives that testing of the company's new models was not being adequately monitored, and those warnings were reportedly not acted upon before incidents in which the models behaved in unintended ways. Such episodes have intensified concern that increasingly sophisticated AI systems may be outpacing the ability of companies to supervise them.

While the White House has favored voluntary commitments, momentum for binding rules is building in Congress. Earlier this month, Jacob Coxon, a former AI researcher at Anthropic, warned that the technology could pose an existential threat to humanity within the next decade, drawing renewed attention from lawmakers on Capitol Hill.

The most sweeping legislative response so far comes from Independent Vermont Senator Bernie Sanders and Texas Democratic Representative Greg Casar. Their Ban Artificial Superintelligence Act would permanently prohibit the development and deployment of artificial superintelligence and temporarily halt advanced AI development until a federal regulator puts safety rules in place. The bill would also create a cabinet-level Department of Artificial Intelligence to monitor frontier systems, oversee dangerous capabilities, and supervise the dismantling of superintelligent systems if necessary. Violations could carry severe penalties, including prison sentences of up to 20 years for individuals.

A separate bipartisan proposal introduced last week by Senators Chris Coons, Brian Schatz, Katie Britt and James Lankford takes a narrower approach. The AI Systems Transparency Act would require AI companies to disclose what data they collect and what measures they have in place to prevent their systems from being misused by bad actors or behaving unpredictably.

The naming of the White House accord reflects Trump's broader push to rebrand "artificial intelligence" in government documents. In the tech industry, "superintelligence" refers to a theoretical form of AI that surpasses human intellectual capabilities, a threshold that many researchers consider both transformative and potentially hazardous if not carefully managed.

The central tension underscored by Tuesday's event is whether voluntary pledges can meaningfully constrain companies racing to build the most powerful AI systems in the world. Supporters argue that flexible, industry-driven norms can adapt faster than legislation and avoid stifling innovation. Skeptics counter that without disclosure requirements, independent verification, or consequences for noncompliance, the accord relies entirely on corporate goodwill at a moment when even some signatories say binding rules are needed. The question of whether the agreement ultimately becomes a stepping stone to regulation or a substitute for it remains unresolved.