Recursive Superintelligence emerges from stealth with $650M raise

A London-based AI startup set up just months ago, which believes it’s pursuing the fastest path to surpassing human intelligence, has today come out of stealth, having raised over $650m at a $4.65bn valuation.

Called Recursive Superintelligence, the funding round was led by GV, Google’s VC arm, and US VC Greycroft, with participation from chip makers Nvidia and AMD. The disclosure of the funding round, follows previous reports about the expected raise.

Recursive Superintelligence’s “bold bet” is that AI systems will improve themselves by analysing their own performance, without human intervention.

The startup's co-founders include Richard Socher, its CEO, who was previously chief scientist at Salesforce, and Tim Rocktäschel, a professor of AI at London’s UCL and a former Google DeepMind scientist. Others who work at the startup, which has a team of less than 30, previously worked at Meta and OpenAI.

The startup, which was incorporated in London and has offices in London and San Francisco, said a clear trend was emerging in AI.

In a blog post on X, it said: "The fastest path to superintelligence will be realised by AI that recursively improves itself, and does so via open-ended algorithms that drive endless innovation.

“We will first focus on the science of AI itself (by creating AI that improves AI), but the playbook we create will soon allow us to revolutionise every scientific discipline. The potential benefits for humanity of safely creating such an advance cannot be overstated.”

Recursive Superintelligence is one of several new AI startups looking at new ways to make improvements in AI intelligence. These include Yann LeCun’s AMI Labs and David Silver’s Ineffable Intelligence.