AI-era unicorns are arriving at a pace that makes older startup playbooks feel like antiques. New analysis from Accel, produced with Dealroom and Revelio Labs and shared with Fortune exclusively, finds that of the 86 new unicorns minted in Europe and Israel from 2023 onwards, 20% reached a $1 billion valuation within two years of founding. That is up from just 5% before the generative AI era.

And it gets sharper the longer you zoom out. Nearly a third of those ultrafast unicorns got there in three years or less, versus 12% previously. Accel also reports that the total number of these ultra-fast unicorns has quadrupled since 2023. Translation: Europe is not just producing more unicorns, it is producing them faster. For founders, investors, and boards, that changes everything about runway assumptions, go-to-market cadence, and how quickly a market can decide who owns it.

So what is driving the speed? Matt Robinson, a partner at Accel and cofounder of GoCardless, points to AI as a general-purpose technology that can be dropped into almost any sector and immediately create value. When the value is clearer, sales cycle times collapse, deal sizes grow, and the time between funding rounds shrinks. But there is an extra accelerant that matters even more than marketing speed: in AI, the first company to meaningfully address a given market, whether legal, coding, customer support, or other verticals, tends to lock in that position. Once locked, it is hard for followers to dislodge the leader.

Robinson also frames it as a timing problem with global exposure. “The window to do that is short,” he said, and it is open everywhere at once. In practical terms, that means competition is not just local anymore. If you are building in Europe but your rivals are in the US, your customers, engineers, and financing networks are already comparing you in the same product category at the same speed.

Another reason ultrafast startups are happening: teams are using their own AI tools to scale. The Accel analysis cites that well-implemented AI can mean smaller teams and less overhead, which helps companies grow faster. It also highlights a “tinkerer” founder profile. Zhenya Loginov, also a partner at Accel, describes the ideal founder in the AI age as someone constantly testing new tools, aware of what is emerging, and ensuring the whole team operates at the same level. This is not just about experimenting. Loginov and Robinson describe founders who are running multiple coding agents at once, automating sales outreach and marketing, and compressing internally what used to take months of engineering time into days.

This shift is showing up not only in speed, but in who is building. Accel’s data indicates that founders of Europe’s post-2023 unicorns are twice as likely to come from Big Tech as their predecessors, 23% versus 11%. They are also twice as likely to hold a doctorate, 18% versus 9%. Academic founders doubled, from 12% to 23%. And the pipeline is changing too: Microsoft and Alphabet have overtaken BCG and McKinsey as the most common pipeline for founders.

The report links that change to the maturation of the European ecosystem, including the arrival of large tech campuses in London, Paris, and Zurich. The idea is that engineers are getting global technology stage experience earlier, then taking that toolkit out to build. The rise in PhD founders is also described as AI-driven, not because every AI company needs a doctorate, but because AI has accelerated breakthroughs in robotics, cybersecurity, and autonomous software. Founders working on those frontiers often come from labs and universities, and the markets they open can be very large. Anton Osika, cofounder and CEO of Lovable, which reached $500 million in annual recurring revenue faster than any European tech company before it, told Fortune that AI has made it possible for anyone to participate in the software economy in ways that were previously closed to them. He argues that the “step change” in who can build is what drives the pace and scale of new companies.

Now zoom out from venture outcomes to regulation and operations, because the policy environment is also shaping how quickly AI models get into the wild. On Friday, the Trump administration lifted its two-week block on Anthropic’s Mythos 5 model, clearing it for release to more than 100 U.S. institutions, including major companies and government agencies. The public-facing version, Fable 5, remains blocked, with talks over reinstatement ongoing. In parallel, the Trump administration asked OpenAI to limit release of its GPT-5.6 lineup, comprising three models, Sol, Terra, and Luna, to a small group of government-approved partners, citing advanced cybersecurity capabilities. OpenAI complied, while making clear it does not believe “this kind of government access process should become the long-term default.”

For executives, this regulatory contrast matters because it affects deployment timelines, enterprise adoption, and how fast teams can iterate in production settings. Even if your company is not Anthropic or OpenAI, boards should care: model availability can compress or expand time-to-value, which then feeds back into funding speed and competitive advantage. In an environment where Europe is seeing ultrafast unicorn creation accelerate, the strategic stakes are simple. If your product relies on AI infrastructure that can be gated, your moat timeline may be shorter than you assume. If your team can deploy AI quickly and repeatedly, you may be able to build the “first meaningful” position that Robinson says tends to lock in.

The big takeaway from Accel’s numbers is not just that unicorns are multiplying. It is that the clock is running faster. Boards that still evaluate strategy on old timelines are effectively underwriting slower motion than the market is now rewarding.