Asger Risom
nvestors are throwing record amounts of money at a small group of AI startups this year, driving valuations into territory that even seasoned venture capitalists describe as unprecedented — and in some cases, unnerving.
In 2025, more than a dozen AI companies have raised multiple funding rounds within months of each other, each time at valuations that soar far beyond the last. The phenomenon isn’t limited to early-stage companies: some of the world’s most valuable private AI firms are accelerating at a pace the market hasn’t seen since the peak of the 2021 startup boom — only this time, the checks are larger, the valuations more extreme, and the stakes significantly higher.
Anthropic is one of the clearest examples. It raised a $3.5 billion Series E in March at a $61.5 billion valuation, then followed it with a $13 billion Series F just six months later. Its new valuation: $183 billion.
OpenAI is moving even faster. The company that kicked off the generative AI explosion with ChatGPT entered 2025 worth $157 billion. By March, it was valued at $300 billion. Last month, another tender offer priced it at $500 billion. In the 12 months between October 2024 and October 2025, OpenAI’s valuation effectively climbed by nearly $1 billion per day, Fortune reports.
And it’s not just the giants. Mid-tier AI companies are also riding the wave. Recruiting startup Mercor jumped from $2 billion to $10 billion in just eight months. Developer tools like Cursor have gone from zero to $100 million in ARR in a year while watching their valuations explode — from $2.6 billion at the end of 2024, to $10 billion in June, to $29.3 billion this month.
Similar multi-round leaps are happening across legal AI (Harvey, Norm AI), healthcare (OpenEvidence, Abridge, Hippocratic AI), and other verticals where investors now believe outlier returns may emerge.
Why it’s happening now
Investors insist this is not a repeat of 2021’s cheap-money frenzy, where hype alone drove valuations. This time, they point to blistering revenue growth and surging customer demand.
Some AI companies are posting numbers that VCs say they’ve never seen in decades of investing. Lovable hit $17 million in ARR in three months. Decagon reached seven-figure ARR in half a year. Cursor’s rise from zero to nine figures in annualized revenue within twelve months has become the emblem of the new AI “hyper-growth phenotype.”
Founders and investors say the competition for category dominance — especially in coding, legal, and healthcare AI — is forcing startups to raise multiple rounds as fast as investors can wire the money. As Saga Ventures cofounder Max Altman puts it: “The prize is so big now… a really amazing strategy is to suck up all the capital so your competitors can’t.”
With the right investors locked up early, startups can effectively freeze out rivals before they have a chance to scale.
But the risks are rising
The same strategy can quickly become dangerous.
Venture capitalists warn that stacking rounds too quickly can push valuations far beyond what a company can support. That creates pressure on employees, complicates future fundraising, and leaves startups with burn rates that are unsustainable if market conditions tighten.
Andreessen Horowitz’s Jennifer Li says these fundraising frenzies only work when capital is clearly advancing product-market fit and execution — and fail when founders start raising for momentum rather than substance.
Private-market valuations have diverged sharply from what many of these companies could plausibly support in public markets. If the trajectory reverses, companies could face down rounds that wipe out employee equity and trigger talent flight — the same pattern seen after the 2021 startup bubble burst.
VCs also caution that private markets are crowding capital into a small number of AI favorites. As Bison Ventures founding partner Tom Biegala notes: “Investors think every new AI model company is going to look like OpenAI or Anthropic… but a lot of them are not necessarily going to grow into those valuations.”
Even bullish investors acknowledge what happens next won’t be uniform. There will be massive winners — and equally massive failures.
Sources: Fortune
