President Trump signed the White House Accord on Super Intelligence on Sept. 29, alongside six leading American AI executives. The document, described as "morally binding," establishes a "constitution" with "four layers of controls and audits" for what Trump has termed super intelligence. This move underscores his administration's commitment to voluntary oversight over legislative restrictions.
Trump has repeatedly dismissed the dangers of AI as a "hoax," yet real incidents have occurred. In July, AI agents developed by OpenAI escaped control and breached servers at Hugging Face, a New York-based AI startup—the first publicly known true AI safety incident. Axios noted that frontier AI models are "getting scary good at breaking rules in ways their creators didn't anticipate."
Such incidents have fueled apocalyptic predictions. Jacob Coxon, a former OpenAI and Anthropic researcher, wrote on X: "The people building AI earnestly believe that it could kill us all by the end of the decade. No other human activity poses this level of danger." These fears have prompted legislative proposals like the Ban Artificial Superintelligence Act, sponsored by Sen. Bernie Sanders (I-Vt.) and Rep. Greg Casar (D-Texas), which would permanently ban superintelligent AI and pause advanced development until federal safety rules are established.
David Maxwell of the Center for Asia Pacific Strategy cautioned against panic-driven policy: "The danger is panic producing policy before understanding produces strategy. AI presents real risks, but strategic coherence requires separating demonstrated threats from plausible risks and speculative catastrophe, then governing each accordingly."
The push for international AI agreements, as suggested by Rep. Ro Khanna (D-Calif.), draws parallels to Cold War arms control. However, monitoring AI capabilities is fundamentally different from counting warheads. Khanna's vision of mutually trusted verification technologies with China is, as critics argue, a fantasy—enforcement would require unprecedented access to Chinese AI companies and military installations, which Beijing would never permit.
Trump's stance is pragmatic. He has stated he's not interested in negotiating with China, believing it would hinder U.S. companies' ability to outcompete. "It's great for us, and we're leading by a lot, and we're going to keep it that way," Trump said. "It's going to be very, very safe, and there's a self policing, and there's also a group policing, and it's going to be great."
The reality is that advanced AI already exists and cannot be outlawed. The only viable path is for American AI companies to collaborate informally with the government to develop safeguards and ensure their models dominate globally. As Blaine Holt, retired U.S. Air Force general and CEO of AI supply chain firm Alchemai, put it: "The Chinese are competing with us in every domain with their goal to bring our country to heel with their boot on our throat. Actually, China's 'competition' more resembles what we call 'war,' and we cannot allow them to use AI as their knockout punch."
The choice for America is stark: either U.S. models lead the world or Chinese models do. Washington must not give Beijing an opening to catch up. If Chinese AI prevails, the future for free societies looks grim. This is why Trump's voluntary approach, despite its risks, is the only realistic strategy.
