AI Startup Valuations: Instinct Hits $10B 33 Days After $2.5B

A quadrupling in a month

The week's largest U.S. startup financing went to Instinct, a San Francisco company building a personal AI agent for everyday tasks. It raised a $1 billion Series C at a $10 billion valuation, with Sequoia Capital, Benchmark and Coatue backing the deal. 15 The timing is what stands out. The round closed 33 days after a $250 million Series B that valued Instinct at $2.5 billion. Index Ventures and Benchmark co-led that earlier round in August. 1

Accounts differ on how the new round was structured. Tech Funding News describes it as led by the three firms. 1 Another account notes that the company's own announcement named no lead investor. 2 By Tech Funding News's tally, total funding now stands at roughly $1.35 billion. 1

Instinct's founder, Noah Shinn, is 23. He was previously a research scientist at Sierra and was first author on a 2023 paper about language agents that learn from failed attempts. 1 He founded the company in 2025, and Crunchbase calls it a year-old company. 15

The investors were buying in without much to go on. Instinct is still in early access and has not shared user numbers, revenue or growth metrics. 12 The one notable change since the Series B is a rewritten privacy policy. TechCrunch had criticized the earlier version in August. 2

Not a one-off

Instinct is not alone in repricing this fast. General Intuition trains AI models on video-game footage and virtual worlds to build skills that carry over to physical environments. 35 It closed a $220 million round on September 29 at a $6.2 billion valuation. Valor Equity Partners and Atreides Management led the deal. 35 That price is nearly triple the $2.3 billion set in June, when Khosla Ventures led a $320 million Series A. Total disclosed funding now exceeds $670 million. 3

The rest of the week's top rounds followed the same pattern. EliseAI, which applies AI to home rentals and is expanding into healthcare, raised $350 million at a $4 billion valuation from Andreessen Horowitz and Bessemer Venture Partners. 5 Armadin, an AI cybersecurity startup, also made the list. Crunchbase summarized the top 10 as almost entirely about AI. 5

The market backdrop

These deals sit inside a funding environment that has shifted heavily toward AI. PitchBook counts $255.5 billion in AI startup funding in the first quarter of 2026. That already exceeds the roughly $254 billion raised in all of 2025. 4 Crunchbase measures the quarter differently, at $242 billion. That figure is 80% of the $297 billion in global venture funding, up from AI's 55% share a year earlier. 4

The headline share hides how concentrated the money is. OpenAI, Anthropic and xAI took about two-thirds of the capital. Robotics, defense tech, vertical AI and developer tools split the rest. 4 Defense tech is still growing, with $14.6 billion so far in 2026 against $9.6 billion for all of 2025. 4

Reading the repricing

The main story here is the gap between price and proof. At most companies, a fourfold valuation jump means some milestone was hit: a revenue figure, a usage curve or a launched product. Instinct's public record over those 33 days shows a privacy policy update and continued early access. 12 General Intuition's near-tripling came over three months, also without disclosed commercial metrics. 3

The likeliest explanation is that investors are not pricing these companies on fundamentals at all. They seem to be pricing scarcity: access to a small number of founders and research teams believed capable of building agent-level or world-model products. Shinn's research background and the willingness of top firms to return within weeks point that way. Benchmark is in both of Instinct's last two rounds. 1

The concentration data supports that reading. 4 When two-thirds of AI dollars go to three frontier labs, funds that cannot get into those deals may be competing hard for the next tier of perceived contenders. That competition pushes prices up faster than any operating data can justify or contradict.

The risk is easy to see. Valuations set this quickly, on this little public information, leave later investors, employees holding equity and eventual acquirers relying on trust rather than evidence. Prior rounds act as the only benchmarks, and they are weeks old. If disclosure keeps lagging this far behind pricing, the first real test of these numbers may be a down round or an exit, not a growth report.

It is still possible that Instinct's early-access users are growing fast and General Intuition's models are working. But the evidence the market has to go on right now is the size of the checks, not what the companies have shown.