Trump's New AI Task Force Has No Budget, No Law Behind It, and an Internal Contradiction

President Trump's Super Intelligence Force, announced October 4, arrives with four named leaders, a new vocabulary mandate, and no statutory basis at all. Unlike the Space Force, which Congress created through a provision in the 2020 National Defense Authorization Act, the SIF exists only because the president said so. It has no enabling legislation, no appropriated budget, and no fixed slot in the federal org chart, a gap that critics writing for The Conversation have already flagged as a structural weakness: without legal footing, the task force's authority over companies or agencies is unclear.

The task force follows two documents signed September 29 at a White House summit. The first is an executive order, "Inaugurating the Era of Super Intelligence," which carries real legal force inside the government: it directs every executive agency to strike "artificial intelligence" and "AI" from official communications and use "super intelligence" and "SI" instead. That is a binding instruction to federal employees, even though it does not touch how any outside company builds or sells its products.

The second document, an "Accord on Super Intelligence," is the one that was supposed to bind industry. Six executives, including Sundar Pichai of Google, Dario Amodei of Anthropic, Mark Zuckerberg of Meta, Greg Brockman of OpenAI, Elon Musk of xAI, and Jensen Huang of Nvidia, signed it the same day, committing to four layers of controls on frontier models. It is voluntary. The Council on Foreign Relations has already described the accord as toothless, noting it names no auditors, sets no incident reporting duty, and attaches no penalty for breaking it.

That leaves the SIF itself as the only enforcement vehicle in the picture, and its composition points to contracts and personnel rules rather than regulation. Jay Clayton, the Director of National Intelligence and a former SEC chair under Trump, chairs the group. Andrew Ferguson, chair of the FTC, and Scott Kupor, director of the Office of Personnel Management, serve as vice chairs. OPM's presence is the odd one: a personnel agency has no obvious role in AI policy unless the administration intends to use federal hiring and contracting rules, rather than new law, as its lever over the industry.

The fourth leader is where the contradiction sits. Emil Michael, Undersecretary of War for Research and Engineering and a former Uber chief business officer, is also a vice chair. Yahoo News has reported separately that Michael is leading the Pentagon's fight against Anthropic over the company's model-use restrictions for defense work. Anthropic is simultaneously one of the six signatories to the administration's own voluntary safety accord. The same administration is treating the company as a project partner and a target inside the same six-day span, with no public explanation of how those two postures coexist.

None of this changes a single statute, regulation, or contract clause. The executive order renames things inside the government. The accord asks companies to police themselves. The task force, for now, is four officials with overlapping jurisdictions and no budget line. What it does change is the political register: the rebrand lands two weeks after Verasight found 63% of Americans want AI development slowed against 5% who want it accelerated, and after CNBC reported Trump sinking on the issue ahead of the midterms. The actual legal fight over AI policy, a year-old and still working through the courts, continues on a separate track entirely.