The Justice Department ordered all employees to replace “artificial intelligence” with “super intelligence” in official documents, court filings, and public communications. The directive marks the latest step in an aggressive federal push to reshape how Americans talk about the technology—and what they expect from it.
Acting Deputy Attorney General Trent McCotter issued the directive on Tuesday, telling staff to use the new terminology “to the maximum extent permitted by law.” The memo follows President Donald Trump’s executive order from last week, which mandated the same change across federal agencies. Trump says the new wording better captures the technology’s potential, though critics worry the rebranding sidesteps rather than addresses growing public concerns about AI’s risks.
The Reframing Without the Oversight
Public confidence in artificial intelligence has eroded as hacking incidents and safety questions have prompted widespread calls for stronger guardrails. Yet Trump has explicitly resisted new regulations for the technology, telling officials the Justice Department would “rein it in if necessary”—a vague reassurance that leaves details to executive discretion rather than statutory protections.
The administration’s approach differs sharply from what many democracies are pursuing. The European Union’s AI Act, for instance, established binding regulations before the technology achieved widespread deployment. The U.S., by comparison, is choosing another path: linguistic reframing paired with minimal legal constraint.
A New Power Structure for AI Oversight
Trump has also created a “Super Intelligence Force,” led by Jay Clayton, who heads the Office of the Director of National Intelligence. Pentagon Chief Technology Officer Emil Michael, Federal Trade Commission Chair Andrew Ferguson, and Scott Kupor, director of the Office of Personnel Management, serve as vice chairs. The group will explore the technology’s benefits and risks while ensuring American competitiveness against China—objectives that don’t always align when corporate interests and public safety diverge.
This concentration of AI governance within the executive branch, without congressional involvement or public regulatory frameworks, raises questions about who truly shapes the technology’s development. The FTC, traditionally a check on monopolistic corporate behavior, now shares oversight responsibility with military and intelligence officials whose priorities center on national competition rather than consumer protection.
Private Sector Racing Ahead
Federal officials are rebranding and reorganizing, while private companies accelerate development with minimal constraint. Reflection AI, a startup backed by Nvidia, released its first open-weight model, Beam, on Monday. The company claims the model rivals leading open-source models developed in the United States and China, including Chinese AI startup Z.ai’s GLM-5.2 and Alibaba’s Qwen3.8-Max.
Developers worldwide can now build on Reflection AI’s technology without oversight from the original creators. In one sense, this democratizes innovation: no single company can control the technology. But accountability also fragments. When an open-source model causes harm, determining responsibility becomes difficult. Who bears liability—the original developer, the company that fine-tuned it, or the user who deployed it? Federal law hasn’t caught up to these questions.
Why This Matters:
The administration’s rebranding and governance restructuring reveal a particular vision of how technology should develop: through corporate and military leadership, with minimal public regulation or legislative oversight. This model prioritizes speed and competitive advantage over the distributed protections that governed earlier transformative technologies, from nuclear energy to pharmaceuticals to financial derivatives. Public pressure and documented harms came before each of those regulatory frameworks.
AI development is advancing faster than any of its predecessors. By the time documented harms accumulate and demand for regulation grows, the technology may already be deeply embedded in criminal justice, hiring, lending, and healthcare systems. Governing through terminology and executive committees rather than transparent statutory frameworks shifts power away from elected representatives and toward appointees answerable primarily to the president. For communities already harmed by automated bias in policing and hiring, the lack of new legal guardrails means existing vulnerabilities will persist—now without even the vocabulary to discuss them clearly.