Guozhen AIGlobal AI field notes and model intelligence

Realtime AI News

Daniel Ek backs AI researcher’s fusion startup Europa Fusion, which bets on simulation over hardware

Bloomberg first reported that Spotify founder Daniel Ek is backing Europa Fusion, a stealth U.K. fusion startup founded by former DeepMind and Pacific Fusion machine-learning researcher Nantas Nardelli. The company is pursuing a pulsed-power design with a simulation-heavy approach, a bet that matters as AI’s electricity demand renews interest in carbon-free power.

Published
Daniel Ek投资AI研究员创办的聚变公司Europa Fusion,走仿真优先路线
Image source: techcrunch.com

Bloomberg first reported that Spotify founder Daniel Ek is backing Europa Fusion, a U.K.-based fusion startup that has kept a low profile. TechCrunch then reported, citing people familiar with the company, what Europa Fusion is actually building and why its approach differs from that of its peers.

The company was founded in February. Its founder, Nantas Nardelli, is an AI researcher, and sources told TechCrunch he had been working on the idea for longer than the company has existed. No investment figure was disclosed in either report.

Nardelli’s career sits almost entirely on the AI side: he worked at Google DeepMind, Facebook AI Research and Microsoft Research Cambridge, and most recently was head of machine learning at Pacific Fusion, a Fremont, California-based fusion startup that recently broke ground on a demonstration facility in New Mexico.

That is where the disagreement starts. Most fusion startups of Pacific Fusion’s size lean heavily on physical devices to test theories and sharpen simulations. People close to Nardelli’s thinking told TechCrunch he did not embrace that hardware-centric focus, arguing instead that fusion companies should invest more in simulation- and data-heavy work.

On physics, though, Europa Fusion follows the same broad path: pulsed power. Such reactors drive precisely timed, enormous surges of electrical current into a fuel target, compressing it until the atoms inside fuse and release energy. The approach puts severe demands on pulsed power supplies, target fabrication and timing control.

Fusion’s appeal is that it could deliver carbon-free electricity from a nearly limitless fuel supply, yet no one has built a plant that supplies power to the grid. For the AI industry the stakes are more immediate: data-center electricity demand is climbing, and any technology that could ease concerns about AI’s energy consumption attracts fresh attention.

Nardelli framed the pivot himself on his website: “My original background is in electronics and control engineering,” he wrote. “I used to build robots for a living. One day I’ll be back to that, but it’s time to fix climate change for good.”

Europa Fusion still has to prove that a simulation-first strategy buys speed on a path whose physics is understood but whose engineering has never closed. Daniel Ek’s name brings attention; fusion timelines still demand results, and the sector’s patience for slippage is finite.

Why it matters

For fusion, the backing gives a stealth startup a high-profile validator; for AI, it points the energy debate toward supply-side solutions rather than efficiency alone.

FusionStartupsDaniel EkAI energy demand
Back to realtime news

Nearby Updates

All

11/21, 04:04

RYPTASK Launches Managed AI Agent Solution Aimed at MSPs

RYPTASK has launched a managed AI agent solution built for managed service providers and their clients, according to an announcement carried by EIN News. The notice does not disclose which models it runs, how it is priced, or when it becomes generally available, leaving the resellable agent product as the confirmed direction.

11/21, 03:32

AI claimed a solution to the 167-year-old Riemann Hypothesis, and a human did too

An AI system has claimed a solution to the 167-year-old Riemann Hypothesis while a human mathematician advanced a competing solution, TechCrunch reports, leaving mathematicians stunned. Neither argument's content, authorship, nor verification status was disclosed, so there is no basis yet for treating the conjecture as resolved.

11/21, 04:49

Elvis Presley's estate sues Anthropic for $46.6 million

The estate of Elvis Presley has sued Anthropic for $46.6 million, according to Wisevoter, adding to the wave of copyright disputes over generative AI training data and outputs. The report does not describe the specific claims, the works involved, or Anthropic's response.

11/21, 02:15

METR's updated time horizon plot: exponential gains continue, judgment lags

METR has published an updated version of its widely cited and widely misread time horizon plot, adding longer and more open-ended tasks to probe how models handle work that cannot be scored automatically. The report finds capabilities on verifiable tasks still improving exponentially, while models remain below human level on judgment-heavy work such as designing evaluations or scoring philosophical arguments.