Recursive Intelligence, a startup that builds an AI system to automatically design and improve AI chips, has raised $300 million at a $4 billion valuation. The company said Monday the round was driven by Lightspeed.
Recursive says the system will be able to create its own silicon substrate layer and accelerate improvements to the AI chip. Rinse and repeat to get to AGI, the say the founders.
The Series A comes just two months after the company officially launched with a seed investment led by Sequoia. It has raised a total of $335 million, the New York Times reports.
Recursive was founded by former Google research CEO Anna Goldie and CTO Azalia Mirhoseini. Their work on a new reinforcement learning method for chip layout design, it is called AlphaChipit’s been used in four generations of Google’s TPU chip, the startup says.
DST Global, NVIDIA’s venture capital arm NVentures, Felicis Ventures, 49 Palms Ventures and Radical AI are also investors.
Recursive is not to be confused with the similarly named startup Recursive, reportedly founded by noted natural language processing neural network researcher Richard Socher. That Recursive is also in talks to raise a massive round at a $4 billion valuation, Bloomberg reported last week. And he also works on AI systems that improve.
And these two aren’t the only new startups working on the concept. As TechCrunch previously reported, Naveen Rao’s new AI hardware startup, called Unconventional AI, is also working on a smart substrate. In December it raised a $475 million seed round at a $4.5 billion valuation led by Andreessen Horowitz and Lightspeed Ventures, with participation from Lux Capital and DCVC.
Techcrunch event
San Francisco
|
13-15 October 2026
