Cognition may be eyeing a $40B valuation

Cognition is back in fundraising talks just three months after its last huge round, and the numbers are already getting bigger.

Cognition is reportedly discussing a new round that could value the Devin maker at $40 billion.

From a $26 billion round to a possible $40 billion jump

The headline here is the speed. TechCrunch reports that Cognition is already in talks for another round after raising $1 billion in May at a $26 billion valuation. If the new deal lands where Bloomberg’s sources suggest, the company would be pricing itself at at least $40 billion in just a few months.

That kind of jump only happens when investors think growth is still running hot. It also tells you something about how much money is chasing AI coding tools right now. The market is rewarding companies that can show real usage, fast revenue growth, and clear enterprise demand.

• $1 billion raised in May 2026
• $26 billion valuation in the last announced round
• Potential $40 billion valuation in the new talks
• Only about three months between rounds

Devin is selling speed, not replacement

Cognition’s product is Devin, an AI coding agent built to handle software tasks that usually eat up engineer time. Scott Wu, Cognition’s co-founder and CEO, told TechCrunch that the company is not pitching Devin as a human replacement. Instead, he described it as a system for the duller work programmers often avoid, including updating old code and moving apps between platforms.

That positioning matters. A lot of AI coding products promise a magical future where engineers disappear. Cognition’s pitch is narrower and, frankly, more believable: let the agent do the repetitive work while humans handle the judgment calls. That makes the product easier to sell inside companies that already have engineering teams and a pile of legacy code.

“Devin is not being sold as a human replacement.” — Scott Wu, Cognition co-founder and CEO, in comments to TechCrunch

Wu also said in May that Cognition had reached a $492 million annualized revenue run rate and that enterprise usage of Devin had been growing 50% month over month for six months. Those are the kinds of numbers that make a new valuation story possible, because they suggest the company is not relying on hype alone.

Why enterprise buyers matter more than demo videos

AI coding tools can look impressive in a demo and still fail in production. What matters is whether companies trust them with real work, and Cognition says its customers include Mercedes-Benz, NASA, and Goldman Sachs. Those names are useful for more than marketing. They signal that the product has moved beyond hobbyist curiosity and into enterprise procurement.

That customer mix also explains the company’s valuation pressure. Big enterprises buy in slowly, but once they trust a tool, they can spend heavily and expand usage across teams. If Cognition can keep showing revenue growth while proving Devin is useful for migration work, maintenance, and other unglamorous tasks, investors will keep treating it like a premium asset.

• Mercedes-Benz, NASA, and Goldman Sachs are listed customers
• Devin is aimed at long-tail engineering work, not flashy one-off coding
• Enterprise adoption can turn into larger, stickier contracts

What this says about AI coding right now

Cognition’s fundraising talk is part of a bigger pattern in AI. The companies getting the largest checks are the ones that can tie model capability to revenue, customer adoption, or both. In this case, the valuation story is being driven by a product that people can understand in one sentence: an agent that does coding chores humans would rather skip.

That is also why this round matters beyond Cognition. If the company can jump from $26 billion to $40 billion so quickly, it sets a new reference point for other AI coding startups trying to raise money on growth alone. It raises the bar for proof, too. Investors will want to know whether those usage numbers keep climbing, whether enterprise customers expand their spend, and whether Devin can keep working on the boring tasks that actually save teams time.

The next question is simple: can Cognition keep turning revenue growth into valuation growth at this pace, or is this the point where the market asks for harder proof?