Random Image Display on Page Reload

This Startup Is All In on Recursive Self-Improvement. Its Latest Model Is Scary Good

This Startup Is All In on Recursive Self-Improvement. Its Latest Model Is Scary Good

Recursive makes the case that the benefits of using AI to build AI outweigh the risks.

A brain shape filled with gears and a plant at several growth stages
Photo-illustration: WIRED Staff; Getty Images

For the past few days I’ve been testing an impressive open-weight AIcoding model called Flint. It suggests that fully recursive self-improvement, or RSI—an exciting and controversial approach that uses AI to build AI—is getting closer.

Flint is the first offering from Recursive, a company that aims to use RSI to advance both AI itself and other areas of science using AI. “Nearly every aspect of the entire model pipeline was actually done and improved upon by AI,” Recursive CEO and cofounder Richard Socher told me.

Socher says his company used its recursive AI system, dubbed the Eureka Machine, to automate most elements of the model’s development. This enabled a company with a handful of researchers to build a world-class model in just a few months, he says.

Socher told me that the Eureka Machine devised a number of innovations, including a new way of training the base model, new model architectures, and code optimized for specific silicon chips. Recursive shared benchmark scores that suggest Flint outperforms other models of similar size and architecture on coding tasks.

AI is already widely used to help build, test, and improve AI thanks to its amazing coding skills. This can speed up the process of refining a model and testing new ideas without the need for as much input from humans.

It is also possible to create a loop through which AI decides what improvements to make and then begins another round of improvement. For the moment this technique is limited, but if AI can contribute to every aspect of model development, then the process could theoretically continue indefinitely.

Some experts worry that letting AI improve itself in this manner could ultimately lead to disaster. These doomsayers claim that AI capable of improving indefinitely could escape human control and might decide that humans are an impediment to its goals.

Those concerns gained wider notice this September when a researcher quit Anthropic and warned that advances in AI, including recursive self-improvement, could lead to human extinction within a few years—a concern then echoed by a number of other experts. The idea of RSI has become so controversial that several bills in Congress propose to ban it.

Recursive sees these fears as more sci-fi than reality and believes legislation to prevent RSI is an unnecessary impediment to progress.

Socher says Flint was built using a system incapable of escaping human control. He says the goal is for the Eureka Machine itself to achieve full recursive self-improvement, but this would still be overseen by humans who could stop the process if it goes awry. “We hope that this shows people that indeed we have all the steps in place” to build such a system, he says.

Recursive has raised $650 million at a $4.65 billion valuation from Google Ventures, Greycroft, Nvidia, and AMD Ventures. Socher was previously the chief scientist at Salesforce, and his cofounders include Jeff Clune, a professor at the University of British Columbia who researches ways of building AI systems capable of learning indefinitely.

I must say I am impressed with Flint. I tested it on a Mac with 48 GB of RAM—a healthy amount but only enough to run a modest-size AI model like Flint, which has 40 billion parameters. The biggest models out there have several trillions of parameters, and they can only run on racks of industrial servers.

I used Flint to build a bunch of programs and found it to be both fast and very capable. In my experience, models of this size can be annoyingly error-prone or slow compared to Claude or Codex, but Flint nailed my coding requests every time. I used it to create a command line to-do list, then tried having it build a program that scours the web for interesting research related to recursive self-improvement. I also had Flint whip up a simple website that explains how RSI works. You can see that here.

It’s unclear how much progress other US labs have made chasing recursive self-improvement, but some published work suggests fully recursive systems are already feasible. A team from Microsoft, Carnegie Mellon University, and the University of Virginia recently published work, for instance, showing how an AI coding agent can recursively improve itself by modifying the software harness that turns reasoning into useful action and its own model’s weights.

Perhaps there are risks to full RSI if you are close to building a godlike intelligence, but for the time being it seems like a very promising way to improve your own AI models. This is something I’ve dabbled with in previous editions of AI Lab.

Indeed, Socher says the approach used to build Flint could make cutting-edge AI more accessible. “It's the only way this very small team can produce a model that's actually better than any other model in its class,” he told me. “This will hopefully show people that [recursive self-improvement] is actually quite nice and will benefit a lot of people.”


This is an edition ofWill Knight’sAI Lab newsletter. Read previous newslettershere.

Comments

Back to top

You Might Also Like

Will Knight is a senior writer for WIRED, covering artificial intelligence. He writes the AI Lab newsletter, a weekly dispatch from beyond the cutting edge of AI—sign up here. He was previously a senior editor at MIT Technology Review, where he wrote about fundamental advances in AI and China’s AI … Read More
Senior Writer

Read More

The Anthropic S-1 Filing Is Also an AI Doomsday Primer

Anthropic thinks “closing the loop” is the best path to AI triumph—and a recipe for “existential” threats to humanity.
Brian Barrett

These AI Experts Want to Do High-Stakes Research Out in the Open

Many frontier labs keep their risky research locked away. Trillium Labs wants to show off its work when it comes to self-improvement and model behavior.
Will Knight

Muse Wonder Puts Meta Back at the AI Frontier

Meta’s biggest model rivals the world’s best—and includes new features to keep agents from going off the rails.
Will Knight

Mistral Says Its New AI Model ‘Le Chonk’ Is the Best Open-Weight Offering Outside of China

With the release of its new trillion-parameter model, Mistral is hoping to demonstrate it’s “still in the race” to build frontier-level artificial intelligence.
Joel Khalili

AI Came for Math. Now It’s Chasing the Rest of Science

AI may soon be the default way to solve theoretical problems in economics, high-energy physics, and other fields, MIT researchers tell WIRED.
Will Knight

Solving Math’s Greatest Problems Was an Art Form. Then Came AI

Mathematics has been one of humanity’s most creative endeavors, akin to painting and poetry. Now, mathematicians are trying to save it from the brute force of AI.
Sophia Chen

Timnit Gebru Believes There Is No ‘Existential Threat’ From AI

One of AI’s fiercest critics believes the doom talk is about founders making money, not saving humanity.
Lauren Goode

Anthropic Says It Will Revolutionize Biology—Some Day

In breathless press releases and posts, the company has touted its models’ ability to deliver scientific breakthroughs. But its IPO filing taps the brakes on just how fast its AI will lead to big discoveries.
Emily Mullin

OpenAI Is Pissing Off a Bunch of Mathematicians—Again

“There’s a perception of mobster behavior” from leading AI companies, one mathematician tells WIRED as OpenAI prepares to release more than 100 new solutions to unsolved problems.
Isabella Ward

The Battle to Be Your Personal AI Agent Is Here

OpenAI's Dots and Meta's Muse are vying to be your AI agent of choice. I've tried both—and I suspect you will too.
Reece Rogers

Kevin Roose Didn’t Use AI to Write His Book About AI

The author of The AGI Chronicles says his book “was written by a very tired, overworked, under-slept human being.”
Katie Drummond

OpenAI Delays Release of Latest Model Over Safety Concerns

The company said its latest Astra model would undergo more work to meet safety standards, and issued an apology for the way it handled the hacking of an Australian government website.
Isabella Ward

*****
Credit belongs to : www.wired.com

Check Also

PLDT and Smart Bring Mobile Connectivity to Remote Villages in Cebu, Bukidnon and Cotabato

PLDT and Smart Bring Mobile Connectivity to Remote Villages in Cebu, Bukidnon and Cotabato

For communities far from major urban centers, reliable mobile connectivity makes a meaningful difference in …