In another episode of Silicon Valley’s favorite pastime — throwing a few billion dollars at an idea that sounds somewhere between genius and Bond villain — Elon Musk’s artificial intelligence startup, xAI, has raised a modest $6 billion in Series B funding. The company, casually aiming to build what it calls “truth-seeking” AI, announced the deliriously large investment on Sunday evening as though it were announcing a new toaster design.
Musk, ever the straight-faced disruptor, founded xAI last July because what the world clearly needed was another AI startup helmed by a billionaire with strong Twitter opinions and a highly selective relationship with “truth.” The company says the new funds will be used to bring its “first products to market,” build out advanced infrastructure and pursue what it vaguely describes as “research and development” — a phrase that here might cover anything from machine learning breakthroughs to hiring more interns who know how to prompt ChatGPT.
The funding round was led by investors who presumably made eye contact with Musk once at a corporate brunch and decided that was enough. Backers include Andreessen Horowitz, Sequoia Capital, Fidelity Management and Valor Equity Partners. None of them elaborated on exactly what they think “truth-seeking AI” means, but one imagines it includes lots of servers and even more philosophical hand-waving.
xAI sits adjacent to Musk’s other unpredictably piloted venture, X (formerly Twitter), suggesting that some integration between the two may occur. Which makes perfect sense, because when you want to build an AI trained on reality and clarity of thought, the obvious first step is tangent (or perhaps TikTok dance) to a social network known primarily for disinformation, meltdowns and extended arguments in all caps.
The company’s main product so far is a chatbot named Grok, which is integrated into X and advertised as having a “rebellious streak” as well as more “humor” than its mainstream rivals. In other words, it was carefully trained to reflect Musk’s own internet personality — a mixture of tech optimism and Reddit comments at 3 am.
Despite the swagger, Grok is still playing catch-up in what has become a fiercely competitive AI race against the likes of ChatGPT from OpenAI, Claude from Anthropic and Gemini by Google, all of whom are desperately trying to create the friendliest-sounding name for something that might accidentally escalate a global conflict one day.
“There will be more to announce in the coming weeks,” Musk posted on X, stern-faced as ever. “Stay tuned.”
Given that previous Musk initiatives promised Mars colonies, self-driving robotaxis and Twitter as a bastion of free speech, we can only assume this latest project will either save humanity or gently rearrange our digital furniture.
After all, what’s six billion dollars between friends when the future of truth itself is on the table?