Mira Murati just closed a $2 billion seed round for her new AI company, Thinking Machines Lab, pushing its valuation to $10 billion, even though the company’s only been alive for six months and hasn’t shared anything about what it’s building.

The deal just wrapped up, and people directly involved reportedly say it’s one of the largest early-stage rounds ever done in Silicon Valley. That’s not just rare. That’s almost unheard of.

The startup, based in San Francisco, hasn’t launched a product, released a prototype, or even given investors a roadmap.

Still, the round was led by Andreessen Horowitz, with backing from Sarah Guo through her firm Conviction Partners, according to the Financial Times. No business model, no demo, no forecast — just Mira and her reputation. And somehow, that was enough.

Thinking Machines hires ex-OpenAI team with no clear product

The company has brought in a long list of familiar names from OpenAI, including John Schulman, who co-founded it, Jonathan Lachman, who led special projects, and former vice presidents Barret Zoph and Lilian Weng. These people worked with Mira at OpenAI before she walked away in September 2023, just weeks before the company tried to kick out CEO Sam Altman.

At OpenAI, Mira led work on ChatGPT, the Dall-E image tool, and the newer voice features that were added right before she left. Before that, she was at Tesla, managing product for the Model X. She didn’t go quietly either.

People close to the leadership crisis at OpenAI allegedly said Mira was one of the executives who questioned Altman’s leadership before the failed board coup in November. For a minute, she was even named interim CEO before Altman returned.

Now, she’s calling all the shots at Thinking Machines. After the new round, she holds voting power on the board that beats every other director combined, giving her full control over all major decisions. That detail came from The Information, which was the first to report the voting structure.

Investors backed Mira’s name with zero product or plan

Despite no product, no pitch deck, and no actual tech revealed, Mira raised billions. She didn’t even pretend to share financials. One investor told Financial Times she didn’t explain the company’s financial plan or business structure during the pitch. Another one who passed said the whole thing was “too secretive.”

Somebody who was pitched said Thinking Machines is probably trying to build artificial general intelligence, the theoretical concept where machines can think and reason like humans, or better. But they also admitted the team is still in the strategy phase, meaning they’re not actually building it yet — they’re still figuring out how.

One investor who went in anyway told Financial Times, “There’s a real finite group of founders, and incredibly smart people. The team [Murati has] pulled together is compelling.” It’s the kind of vague confidence that only exists in a place where $2 billion can be handed out without a product.

The startup did post something online back in February, saying it wanted to make AI “more widely understood, customizable, and generally capable.” That’s the only public statement they’ve made. No site updates since. No launch. Nothing in testing.

But Mira’s not the only one getting funded without showing anything. Ilya Sutskever, another OpenAI co-founder, raised $2 billion in April for a company called Safe Superintelligence, with a $32 billion valuation. Just like Thinking Machines, Sutskever’s startup didn’t have a product either. But unlike Mira, he didn’t get voting rights that let him overrule every single person on the board.

So far, Thinking Machines Lab hasn’t said when it will release anything or what its first project will be. The only thing it has shown the public is Mira’s name, her team of ex-OpenAI talent, and now, a $2 billion runway to figure the rest out. Investors who didn’t join said they needed more details. But the ones who backed it clearly didn’t care.

Cryptopolitan Academy: Coming Soon - A New Way to Earn Passive Income with DeFi in 2025. Learn More