Trump: Rejects Globalist Plans to Regulate AI
U.S. President Donald Trump stated in a speech at the United Nations General Assembly that the U.S. "completely rejects any attempts to create globalist plans to control artificial intelligence" and advocated for the term "super intelligence" instead.
He claimed, "Whoever wins super intelligence, wins," and noted that the U.S. is currently far ahead of China and other countries, and will not stifle growth that is "greater than the Industrial Revolution." He also mentioned that oversight will be very careful, assigned to the Department of Justice and other law enforcement agencies, and can be constrained if necessary, but the policy is to encourage rather than restrict. He further stated that official U.S. documents will use the term super intelligence.
The speech comes ahead of a Security Council meeting on AI and international security on Wednesday, where OpenAI's Sam Altman and Anthropic's Dario Amodei are scheduled to report. Secretary-General Guterres called for multilateral risk management and independent oversight on the same day. Trump previously referred to claims that "AI will destroy humanity" as a scam and specifically opposed calls to slow down certain laboratories.
White House advisors earlier ruled out establishing an AI approval agency similar to the FDA, leaning towards a voluntary framework. Before the visit of the Chinese President to Washington, AI was listed as a potential topic. Trump stated that the Department of Justice has already taken action on related matters.
The stance is a hedge. Buyers are laboratories seeking international testing and common standards, along with initiatives from France, the UK, and Canada, while sellers are U.S. administrative departments keeping rule-making power within domestic law enforcement and industry voluntary compliance. Funding and regulatory expectations are shifting from New York's multilateral venue back to the Department of Justice; the beneficiaries are U.S. training clusters unwilling to accept cross-border assessment barriers, while the pressured parties are companies that have publicly stated their positions on slowing down and independent assessments and are about to speak at the Security Council. The speech does not legislate but sets a U.S. red line for the following day's Security Council meeting.
Source: Public Information
ABAB AI Insight
Trump redefined the upcoming Security Council agenda as a "globalist control plan," substituting the Department of Justice for international assessment bodies and renaming instead of classifying risks. On September 14, he stated that the safeguard is the president himself, specifically naming Anthropic. In the same week, Guterres called for a multilateral framework, Altman sought common standards, and Amodei advocated for slowing down, with three sets of texts colliding in New York. The U.S. excluded an "AI version of the FDA," placing the power of constraint in post-enforcement rather than pre-approval.
The capital pathway prioritizes competition over conventions. The leading narrative is used to block potential international testing clauses that could slow down training and exports. The Department of Justice is framed as having existing tools, meaning that regulatory products are domestic cases, not UN documents. The next day's White House dinner coincides with the Security Council report, with bilateral transactions and multilateral standards competing for the same 48 hours. Policies on electricity and data center costs are handled separately from cross-border security standards.
This aligns with the U.S.'s long-standing refusal to accept supranational jurisdiction like the International Criminal Court and its denial of global carbon taxes on climate issues. The opponent is the Security Council tradition that has incorporated AI into the peace and security agenda since 2023. The industry is in a period of division, with the U.S. defining technology through national competition and laboratories defining technology through security conventions.
Structural changes mean regulatory changes. Rule-making authority is pulled back from international organizations to domestic law enforcement and industry voluntary compliance. The mechanism is: whoever controls the Department of Justice in the country where the training cluster is located decides what constitutes "necessary constraints"; multilateral meetings can listen but cannot set approval gates. Renaming replaces risk language with competition language, making it even harder to write an assessment framework into a binding text.
ABAB News · Cognitive Laws
- Renaming is to replace the definition of risk
- Department of Justice oversight cannot substitute for cross-border testing gates
- Once the competition red line is drawn, the convention becomes optional.