Trump Signs 'Super Intelligence' Executive Order, Tech Giants Sign Voluntary Self-Policing Accord

President Trump signed an executive order Tuesday directing every federal agency to replace the word "AI" with "Super Intelligence" in official communications — a branding shift with no policy teeth. At the same White House summit, executives from OpenAI, Google, Meta, Anthropic, Nvidia, and xAI signed a voluntary accord committing to internal safety controls, outside audits, and board-level review committees, but with no legal enforcement mechanism or disclosure deadlines.

The rebranding and the accord are the twin outputs of the White House AI Summit held September 30, 2026. The executive order itself contains no new regulations, no funding allocations, and no agency directives beyond the terminology change. The voluntary accord, described in the White House readout as "morally binding," similarly carries no statutory weight.

What the executive order actually says

According to NPR's reporting on the [White House](/government/white-house-ai-framework-preempt-state-laws) summit, the EO instructs federal agencies and departments to update official documents, press releases, and public-facing materials to use "Super Intelligence" instead of "AI." No timeline for compliance was specified. No office was named to enforce or track the change.

What the tech accord commits to

The six signatory companies — OpenAI, Google, Meta, Anthropic, Nvidia, and xAI — agreed to three categories of action:

• Internal safety controls: Each company commits to maintaining internal review processes before deploying new models
• Outside audits: Signatories will allow third-party auditors to assess safety practices, though the accord does not name who those auditors will be or how findings will be reported
• Board-level review committees: Each company agrees to establish board committees with oversight of AI safety decisions

None of these commitments include a timeline. None specify what "safety controls" means in practice. No government body was designated to receive audit results.

Why the voluntary approach matters — and why it's limited

The accord follows a pattern set by the Biden administration's 2023 voluntary commitments from the same set of companies. Those commitments, which included red-teaming and watermarking, have seen mixed implementation. Independent researchers have found that watermarking, in particular, remains inconsistently applied across the industry.

The difference this time is the explicit "morally binding" framing. That language signals the White House is leaning into reputational accountability rather than regulatory enforcement — a deliberate choice in an administration that has rolled back Biden-era AI regulations including Executive Order 14110.

For businesses that rely on these six companies' platforms and APIs, the accord's lack of specifics means nothing changes operationally. Enterprise compliance teams cannot cite today's accord as meeting any regulatory threshold in the EU, UK, or California — the three jurisdictions with binding AI rules currently in force.

The "Super Intelligence" terminology shift

Renaming AI "Super Intelligence" in federal communications has no direct policy effect, but it does matter for procurement language, grant applications, and agency RFPs. Any federal contractor or grantee whose documentation references "AI" may eventually need to update language to match federal terminology — a potential compliance paperwork burden for firms with existing federal contracts.

The term "super intelligence" has a specific meaning in AI research circles: it refers to a hypothetical system that surpasses human cognitive ability across all domains. Using it as a synonym for current AI systems — including narrow models that do one or two tasks — is technically inaccurate and could cause confusion in procurement and research contexts.

What to watch

Watch whether federal agencies actually update their terminology and on what timeline — agency inertia on terminology changes is common, and without an enforcement mechanism, the EO may have limited practical effect. More importantly, watch whether the tech accord's audit commitment produces any publicly visible results, or whether it follows the 2023 voluntary commitments into quiet non-enforcement.