Webmasters LDN
Contact
← Blogs

September 30, 2026 · Blogs

Trump Orders US Federal Agencies to Call AI ‘Super Intelligence’

President Trump has told US federal agencies to use ‘Super Intelligence’ instead of ‘artificial intelligence’. What the order does, what it leaves unchanged and why it matters.

On Tuesday 29 September 2026, President Donald Trump signed an executive order telling US federal departments and agencies to replace "artificial intelligence" and "AI" with "Super Intelligence" and "SI". The order covers the language of the US federal executive branch only. It keeps the existing legal definition of the technology and gives no instructions to businesses, US states or other countries.

What the order says

The order is titled "Inaugurating The Era Of Super Intelligence". Its main provisions:

  • Agencies must use "Super Intelligence" and "SI" in official correspondence, public communications, websites, reports, policy documents and other non-statutory documents, as far as the law allows.
  • The executive branch "will not acknowledge" the older terms "in any applicable setting".
  • Agencies do not have to amend regulations, presidential actions, contracts, grants or other documents already issued.
  • For now, "SI" means the technologies covered by the existing definition of artificial intelligence in US federal law (section 9401(3) of title 15 of the US Code).
  • Within 60 days, the Assistant to the President for Science and Technology must send the President draft legislation for a federal definition of "Super Intelligence", with an assessment of whether it should modify, expand on or replace the current legal definition of AI.

The order argues that, as the capabilities of frontier systems improve, they increasingly represent "not merely artificial intelligence, but a new era of Super Intelligence".

How the new name came about

Trump announced the term at the UN General Assembly on 22 September, saying the word "artificial" made the technology "sound fake", the BBC reported. Around the same time, he asked followers on Truth Social to vote on alternatives such as superior, extreme and supreme intelligence.

A week later he hosted technology leaders at a White House lunch and signed the order shortly afterwards. "We've changed the name officially today," he told reporters, according to CNBC. CNBC also reported that the document the executives signed does not declare any change of name, and uses the new term only in its title.

The accord signed the same day

After the lunch, Trump announced that he and six technology leaders had signed a two-page "White House Accord on Super Intelligence", subtitled "Joint Commitment on Frontier Responsibilities". The signatories were Sundar Pichai of Google, Dario Amodei of Anthropic, Mark Zuckerberg of Meta, Greg Brockman of OpenAI, Elon Musk of SpaceX and Jensen Huang of Nvidia.

Each company commits to four layers of controls:

  • internal controls that monitor its models, including checks that they do not "hack or access technical systems in unintended ways"
  • an internal team to make sure those controls work
  • an independent external auditor or evaluator
  • an independent board committee that receives the reports

The commitments are voluntary. Trump described the accord as "morally binding", and the text says it may make sense, over time, to turn the steps into law or regulation. The Guardian reported that it appears to carry no enforcement mechanism and that none of its controls involve government regulators.

Reactions varied. Zuckerberg called the meeting a historic conversation. Kimberlee Weatherall, a law professor at the University of Sydney, told the BBC the accord was "deeply unimpressive" and appeared to let companies decide what counts as safety. Shri Narayanan of the University of Southern California told the Associated Press the accord aims to balance room for innovation with regulation.

What it means in practice

For now, the change is one of wording. Under the order's definition, "SI" and "AI" cover the same systems, and documents already issued can stay as they are.

Among AI researchers, "superintelligence" has a more specific meaning: a hypothetical intellect that surpasses humans in every respect, an idea often linked to the philosopher Nick Bostrom, the BBC noted.

What is still unclear

  • The new definition. The adviser's proposal is due within 60 days, and its content is not yet known. It takes the form of draft legislation, so the definition in US law changes only if Congress passes it.
  • Adoption beyond government. Zuckerberg and Musk have started using the term, and Huang said data centres should be called "super intelligence factories", according to the BBC. Two University of Melbourne academics quoted by the BBC take different views: Simon Coghlan, a digital ethics lecturer, doubted the label would last, partly because it overstates what current AI can do, while Jeannie Paterson said the US might be able to set the terminology now that industry leaders use it.
  • Oversight of the accord. Trump said he would create a board to oversee AI safety but did not say who would sit on it, according to the BBC. The Guardian reported that the accord appears to let companies choose their own evaluators and decide whether to publish the results.

Why it matters to UK businesses

The order does not change UK law or the terms UK organisations use. For businesses that use AI, two points stand out:

  • Working with the US government. If you supply or work with US federal agencies, expect "SI" in their correspondence and documents, and read it as AI.
  • Commitments from major AI companies. If you use tools from the accord's signatories, its four layers of controls set out what those companies have promised. The steps are voluntary, and the accord names no regulator to enforce them.

In the UK, Prime Minister Andy Burnham told the UN General Assembly that Britain would use its 2027 G20 presidency to push for "a single set of global principles and standards" for AI, City AM reported. The paper said the plan sets Britain apart from Trump on how much governments should step in. Separately, MPs have called Google DeepMind, Meta, OpenAI and Anthropic to a hearing on 13 October on whether independent testing of AI models before release should be a legal requirement.

Sources