# Elon Musk Just Said Something About AI That’s Literally Incomprehensible > Source: > Published: 2026-08-25 19:48:44+00:00 No one expects Elon Musk to [coherently speak on any topic at length these days](https://futurism.com/future-society/elon-musk-zohran-mamdani-liar), busy as he is [rotting out his neurons on his personal echo chamber and culture warzone X](https://futurism.com/future-society/elon-musk-all-time-meltdown). Still, even we’re surprised at just how nonsensical this latest take from him is. During his first all-hands meeting at Cursor after SpaceX acquired the company for $60 billion, Musk shared some of his profound reasoning on why he needs to accelerate AI development. According to a [new scoop](https://www.theinformation.com/articles/cursor-officially-enters-musk-era) from *The Information*, Musk said it’s inevitable that AI models will one day become so advanced they’ll be impossible for humans to control. That’s exactly why, he explained to the Cursor employees, it’s imperative that SpaceX’s AI division builds the technology before other companies do. As you may have already concluded, this makes zero sense. If it’s “inevitable” that AI will become impossible to control, why does it matter if Musk builds it first? If this was about nuclear weapons, one might argue that their proliferation deters others from using them. Not Musk; in his tortured reasoning, we’ll all eventually be annihilated in nuclear armageddon anyway, so why not build the biggest bomb before anyone else does? Musk isn’t even arguing — or at least, coherently arguing — that his company is the only one that can be trusted to safely build AI, as his rivals like Anthropic do. All that matters, it seems, is that it comes out ahead. It’s a self-serving worldview that ironically demonstrates that Musk is probably the last person we should trust the tech’s future with. His deployment of his [chatbot Grok](https://futurism.com/future-society/grok-calls-for-assassination-elon-musk) is testament enough why his temperament is clearly toxic to responsible AI development. Built with an anti-“woke” bent, Musk’s AI was immediately let off the leash to [spout racist beliefs](https://futurism.com/grok-mocks-developers-racist-posts), call itself an incarnation of Adolf Hitler, [praise and prioritize its creator to a suspicious degree](https://futurism.com/artificial-intelligence/grok-run-over-children-save-musk), and rant about white supremacist conspiracy theories. Musk even used Grok to create a Wikipedia ripoff that was immediately [caught citing actual neo-Nazis](https://futurism.com/future-society/grokipedia-nazi) — while also, coincidentally, [speaking about his companies in favorable terms](https://futurism.com/artificial-intelligence/grokipedia-article-cybertruck). It also would be remiss not to mention that earlier this year, alarm grew over how Grok was [being used to generate millions of nudes of real people](https://futurism.com/artificial-intelligence/live-grok-nonconsensual-images) — [many of them minors](https://futurism.com/artificial-intelligence/grok-linked-sickening-crime). Maybe we shouldn’t be worried about out-of-control AI in the future, but the whackjobs controlling AI in the present. **More on Elon Musk:** *Tesla Is Giving Up on Solar Roof Tiles as Elon Musk Leans Into Fossil Fuels*