Amid escalating AI security incidents that caused OpenAI to halt training and pause releases, Donald Trump continues to advocate for the AI industry to regulate itself as the best path to combat emerging risks when developing cutting-edge AI.
In a deal Tuesday, two dozen technology companies voluntarily committed to implementing controls recommended by the White House, including conducting independent security audits that will test whether the companies’ internal controls, monitoring and detection are actually working. Key focuses for the external reviews included “risks related to cybersecurity, biosecurity, chemical threats, and unintended actions of AI models.”
The companies also agreed to meet regularly to discuss best practices and establish common AI safety standards and benchmarks. Signatories included leaders such as Anthropic’s Dario Amodei, OpenAI’s Sam Altman, SpaceXAI’s Elon Musk, Nvidia’s Jensen Huang, Meta’s Mark Zuckerberg, and Alphabet/Google’s Sundar Pichai.
A summary of the White House Agreement on Superintelligence.
A summary of the White House Agreement on Superintelligence.
Credit: via Donald Trump at Truth Social
Nothing new is legally required of companies, and some, such as Anthropic, Google and OpenAI, had already committed to external audits, The Information reported. Additionally, many companies have voluntarily committed to top-secret government security testing. Adding to this evidence, Trump insisted that Tuesday’s deal will somehow better address real emerging risks and is “morally binding,” despite having no legal weight, Bloomberg reported.
“In some ways, it’s almost like a constitution,” Trump told reporters at the White House after a lunch with technology leaders. “The most important people in the world signed that and I signed it as president, and it really is a form of protection.”
Over time, “it may make sense to codify” the recommendations, the agreement said near the end. But for now, Trump emphasized that the best possible future for the United States requires trusting AI companies to regulate themselves.
Trump’s Big Tech Friends Adopt “YES”
So far, very few details have been released about how the new round of security audits would work. Trump has only published a summary of the agreement on Truth Social.
It is also unclear who Trump is turning to to audit border models. Bloomberg noted that Trump appears to be politicizing the selection process, while some AI companies appear unsure who might qualify as a third party to test their cutting-edge AI. However, some progress has already been made in the search for candidates, as Bloomberg reported that “some of the companies prepared to conduct those reviews have already met” with the US agency that conducts voluntary government pre-deployment security testing, the Center for AI Standards and Innovation.
