Trump's 'Super Intelligence' Executive Order and the White House SI Accord: What Changed and What It Means for AI Teams
On September 29, 2026, President Trump ordered federal agencies to replace 'artificial intelligence' with 'Super Intelligence' in non-statutory documents, while leading AI companies signed a voluntary White House accord. Here is what the order actually says, what the accord commits to, how it connects to the US-China AI stance, and what engineering teams should do.
Trump's "Super Intelligence" Executive Order and the White House SI Accord
On September 29, 2026, President Donald Trump signed an executive order directing the federal executive branch to stop using the terms "artificial intelligence" and "AI" and to use "Super Intelligence" (SI) instead. The same day, leading technology companies signed a separate, voluntary White House Accord on Super Intelligence after a summit with the President.
The headlines focused on the rename. For people who build and sell AI systems, the more useful questions are narrower: what does the order actually require, what did companies commit to, and does any of it change how you ship software?
What the Executive Order Says
According to the White House fact sheet, the order:
- Directs executive departments and agencies to use "Super Intelligence" and "SI" in official correspondence, public communications, policy documents, and other non-statutory documents.
- Instructs agencies to discontinue the terms "artificial intelligence" and "AI" in those materials.
- Exempts statutory documents. Laws passed by Congress that use "artificial intelligence" are not rewritten by the order.
- Tasks the Assistant to the President for Science and Technology with proposing a federal definition of "Super Intelligence" and identifying any further executive action needed.
The stated rationale is that the old term undersells the technology. The fact sheet says the name "conveys the true capabilities of the technologies being developed today," and the order argues that "Super Intelligence" better captures their "promise, potential, and rapidly advancing capabilities."
What It Does Not Do
It is worth being precise, because a lot of commentary is not:
- It does not create new safety requirements, licensing, or testing mandates for companies.
- It does not change statutes, so legal definitions of AI in existing law remain.
- It does not define "Super Intelligence" yet; that definition is still to be proposed.
- It applies to the federal executive branch's language, not to private companies.
The White House Accord on Super Intelligence
The order followed a White House summit with technology leaders. Reports from CNBC and other outlets named participants including Elon Musk, Meta CEO Mark Zuckerberg, Nvidia CEO Jensen Huang, Anthropic CEO Dario Amodei, Google CEO Sundar Pichai, and OpenAI President Greg Brockman.
The summit produced a voluntary accord. Based on public statements reported by Fox Business:
- House Speaker Mike Johnson described it as "a statement of principles" with "commitments that are voluntary on behalf of the industry."
- Signatories committed to "robust internal controls and layers of internal and external review" for frontier systems.
- Zuckerberg described the commitment as "implementing robust internal controls and multiple layers of audits and reviews."
- Trump called the accord "morally binding."
The important engineering and legal fact: this is self-regulation. There is no enforcement mechanism described in the public materials, and no new federal agency oversight.
The US-China Context
The rename landed days after a three-day summit between Trump and Chinese President Xi Jinping. Afterward, Trump publicly rejected integrating AI efforts with China:
- "I would rather not integrate, because we're leading by a lot," he said, adding that "when you're leading, you don't open it up to each other."
- He rejected a US-China AI "joint venture," saying it could mean giving away "a lot of secrets."
Reporting also noted the two sides agreed to set up a "U.S.-China Super Intelligence (SI) Dialogue" before November 2026 and a bilateral channel for AI-related incidents. So the posture is competition with a narrow communication channel, not a full freeze on contact.
The Washington Post characterized the broader picture as the administration rejecting demands for binding AI rules while Xi called for "human control" of the technology.
What This Means If You Build or Sell AI Systems
For most engineering teams, the practical impact is small but not zero.
1. Federal procurement language will shift
If you sell to US federal agencies, expect solicitations, statements of work, and agency communications to start using "Super Intelligence" or "SI." Your proposals should mirror the customer's terminology while keeping precise technical descriptions. "SI" in a federal RFP will generally mean what the rest of the industry still calls AI.
Action: add "Super Intelligence" and "SI" as synonyms in your capture and proposal keyword lists, and in any contract-analysis tooling that searches for AI clauses.
2. Legal definitions still say "artificial intelligence"
Because statutes are exempt, compliance work tied to laws that define "artificial intelligence" continues to use that term. Do not rewrite compliance mappings based on the rename.
3. Voluntary commitments shape customer expectations
Voluntary accords rarely bind startups, but they set the vocabulary of enterprise due diligence. Expect security questionnaires to ask about:
- Internal controls for model releases
- Independent or external review of high-risk systems
- Incident response for AI-caused failures
If you are building agentic systems for enterprises, having documented evaluation, approval gates, and audit logs is becoming table stakes regardless of regulation.
4. Export and partnership risk with China remains high
The public rejection of US-China AI integration signals that cross-border AI partnerships, model sharing, and compute arrangements involving China will stay politically sensitive. Teams with Chinese partners or open-weight model dependencies should keep supply-chain and licensing reviews current.
Key Takeaways
- The September 29, 2026 order renames AI to "Super Intelligence" (SI) in federal executive-branch, non-statutory documents; statutes are exempt and a formal definition is still pending.
- A separate voluntary White House Accord on Super Intelligence commits signatories to internal controls and internal and external review, with no described enforcement.
- The administration rejected US-China AI integration while agreeing to a limited SI dialogue and incident channel.
- For builders, the impact is mainly terminology in federal procurement and rising expectations for documented controls, not new legal obligations.
Sources
- White House: Fact Sheet on the Era of Super Intelligence (Sept 29, 2026)
- CNBC: Trump tries to rename AI "super intelligence"
- Fox Business: Executive order and the separate SI accord
- Al Jazeera: Trump rejects combining US-China AI efforts
- The Washington Post: Trump rejects demands for AI rules, while Xi calls for "human control"