Can ‘super intelligence’ and a non-binding safety pact solve AI’s image problem?
TechCrunch Anthony Ha ● Covered by 6 sources
Trump made top AI CEOs sign a new safety pledge and renamed AI “super intelligence.” The pact looks flimsy; the branding push looks like the real point.
Based on reporting by TechCrunch, Anthony Ha — read the original for the full story.
Summary, retelling and take written by AI under human oversight; images are AI-generated illustrations. How we work · Report an error
President Donald Trump brought some of the biggest names in AI to the White House this week, but the headline wasn’t just the guest list. The administration is now telling official U.S. representatives to say “super intelligence” instead of artificial intelligence. That shift was folded into a broader push about AI safety, and it landed fast enough to feel improvised.
On TechCrunch’s Equity podcast, the reaction was bluntly skeptical. The new “Joint Commitment on Frontier Responsibilities” looks voluntary to the point of meaning very little, and Sean O’Kane pointed out that Trump called it “morally binding,” which is doing a lot of work for a phrase that has no legal force. The agreement also reportedly contained a misspelling of “United States,” which says something about the speed here.
The politics behind the photo op seem messier than the handshake line suggested. Kirsten Korosec said Anthropic had been in the doghouse not long ago, before CEO Dario Amodei was invited to a 10 p.m. dinner and then showed up again at the luncheon with Mark Zuckerberg, Jeff Bezos, Elon Musk, and other major tech figures. But even with that reset on camera, the relationship inside the administration still looks split.
That split matters more than the pledge itself. Anthony Ha noted that the Treasury Secretary appears friendly toward Anthropic, while the Defense Department remains sharply opposed and has kept calling the company a national security risk. So the clean, unified message from the White House is probably cleaner on paper than in practice. What looks most durable is the rebranding effort: “AI” as scary, job-stealing machine; “super intelligence” as something softer, safer, and easier to sell.
My take — AI-written commentary, not fact-checked reporting
This is classic Washington cosplay with better tailoring. Rename the thing, sign a paper nobody has to obey, then act surprised when the hype machine keeps humming. The real tell is how quickly “AI safety” became a branding exercise instead of a policy one — which is exactly how the industry likes it.
Read more about this at: TechCrunch