Never a fan of bad branding, President Trump has declared war on the term “artificial intelligence” and wants everyone to call it “superintelligence.”Now the masters of Silicon Valley—who count as some of the White House’s most frequent visitors—are being put to the test: Can they get on board with the new messaging?

Unlike the president’s edicts to rename large bodies of water or a mountain in Alaska, rebranding AI is trickier for companies that have staked their futures, even names, on it. But Trump said “artificial” makes it sound fake, which he sees as a problem.“It is not fake. It’s actually amazing,” he told the United Nations General Assembly in late September. Days later, he signed an executive order that all government documents with the wording be renamed.

“The White House considers anyone that uses the term, ‘Artificial Intelligence,’ as opposed to the highly accepted new and more accurate term, ‘Super Intelligence,’ THE ENEMY!”Trump said last month on Truth Social.

Tech executives have been walking a linguistic tightrope ever since, sometimes recalibrating in midsentence. “What should a building code for AI safety—or SI safety—look like?”Microsoft President Brad Smith said on a podcast in October. A Microsoft spokeswoman said the company uses a range of terms for the technology.

Nvidia Chief Executive Jensen Huang, who has a close relationship with Trump, has used SI for certain audiences, and AI for others. “The future of Super Intelligence will be shaped not only by what it can do, but by the confidence it earns,” he posted on X the day after he and other tech leaders signed a pledge at the White House to keep the technology safe.

Huang was back to the old lingo at a conference in Berlin a few weeks later: “Every company in Europe will become an AI company,” he said.

Not one to shy from jettisoning a world-famous brand (remember Twitter?), SpaceX and Tesla CEO Elon Musk took to the new terminology almost immediately. SpaceX’s AI division will become SpaceXSI, he declared. In a securities filing, Tesla said its mentions of SI “refer to the technologies and activities that we previously described as ‘artificial intelligence’ or ‘AI.’”

Salesforce CEO Marc Benioff said he is also getting with the program. The company’s AIForce platform is now SIForce, he said on X: “The era of super intelligence is here.”Making the change is trickier when “AI” is on the shingle. OpenAI CEO Sam Altman has said the company won’t change its name. “Brands are sticky,” he told Politico.

The real test has come in recent days. Between a wave of tech-industry earnings reports, Anthropic’s public-offering filing and last week’s stock-market meltdown over AI, or SI, spending, company bosses have had all sorts of opportunities to embrace—or skip—the rebrand.

Of 119 S&P 500 companies discussing the technology on investor calls, securities filings or news releases since last week, five have used some variation on “superintelligence,” a Wall Street Journal analysis found. Some use one word, others two, and there is a split over whether it is capitalized.

“People are muddling through,” said Brendan Steinhauser, CEO of the Alliance for Secure AI, a nonprofit advocating for AI-safety policies. “They’re typically finding a way to not be oppositional to the White House but also saying things like, ‘We’re going to continue to talk about this in the way we have talked about it.’”For one thing, “superintelligence” was already widely used in AI circles to describe (so far, hypothetical) models more cognitively capable than the most gifted humans. Steinhauser said his group aims to preserve that definition and keep AI as the more general term.

“Yes, we will continue to be called The Alliance for Secure AI,” the group posted on X in October, borrowing Trump’s signature signoff: “Thank you for your attention to this matter.”

Even within Trump’s White House, change is hard. A recent Facebook post from first lady Melania Trump’s account—announcing a computing challenge for students—referred to AI instead of SI in one instance. One commenter gently called out the lapse: “I’m still correcting myself also.”

On the “All-In” podcast co-hosted by Trump’s former AI czar David Sacks, the tech founder pointed out that listeners were scrutinizing when the hosts reverted to “AI” instead of “SI.”A late October episode was titled “AI Cyber Panic.”“You’re getting some shade for that,” Sacks joked with his co-host, tech entrepreneur Jason Calacanis, last week.

At one Washington conference on robotics, a top researcher at the National Institute of Standards and Technology kicked off a panel discussion with a disclaimer about his use of superintelligence.“No question I’m going to slip up at some point,” said Craig Schlenoff, who leads NIST’s AI research, measurement and standards division. “I haven’t completely cleared my cache yet, so muscle memory is going to jump in and AI is going to come in.”