By applying its Uncertainty Engine, AI company DigiLab has helped the UK Atomic Energy Authority (UKAEA) to drastically accelerate time of fusion research, potentially leading to faster route toward clean fusion energy.
“With DigiLab’s Uncertainty Engine, we’ve helped UKAEA create fast, explainable models that don’t just make predictions, but also quantify uncertainty making data actionable, even when it’s sparse,” said Amanda Niedfeldt, Head of Business Development, DigiLab comments on the breakthroughs. “Some examples of the powerful impacts this unlocks are drastically faster simulations, fewer redundant runs, and smarter diagnostic decisions, which accelerates design programmes and prevents costly late-stage redesigns.”
Fusion energy has long been seen as the “holy grail” of clean power, capable of delivering abundant, carbon-free energy. But the science is staggeringly complex. At the heart of the challenge is turbulence – the behaviour of plasma at over 150 million °C that bleeds energy and prevents stable reactions. Predicting turbulence reliably is essential, but traditional computational models demand millions of CPU hours and often fall short.
“In fusion research, the cost of uncertainty is high – in compute, in time, and in design decisions,” said Niedfeldt.
Working with UKAEA’s flagship STEP programme (Spherical Tokamak for Energy Production), DigiLab has delivered results across turbulence simulation and intelligent sensing. For turbulence, advanced machine learning models now predict behaviours in spherical tokamaks, one of the least understood areas in plasma physics, giving scientists powerful tools to design better-performing reactors.
For sensing, DigiLab has applied probabilistic AI with genetic algorithms and Bayesian optimisation to rapidly identify the best sensor configurations for fusion devices. This prevents costly late-stage redesigns and strengthens the resilience of multi-billion-pound fusion assets.
