Holding a plasma in shape.
Researchers used reinforcement learning to control magnetic confinement in a research tokamak.
- Maturity
- Demonstrated, stage 3 of 4
- Support
- 1 source · paper
- Evidence detail
- How we know ↓
Hot plasma must be controlled within a complex magnetic system.
Controllers require specialized engineering for target configurations.
A trained policy commanded the tokamak’s magnetic coils.
What was shown, and what wasn’t
Shown
The controller demonstrated multiple plasma configurations on TCV.
Not shown · limits
- Plasma control is not net energy production or commercial fusion.
Bright editorial interpretation
A tool for researchers exploring fusion reactor designs.
Still open
Test broader conditions and more demanding devices.
How we know1 source · checked 2026-09-05 · no corrections
Original sources
Institutions: EPFL & DeepMind
- Maturity
- Demonstrated
- Source published
- 2022-02-16
- Last source review
- 2026-09-05
- Editorial method
- Original source check
- Place / relevance
- Lausanne, Switzerland · institution-location
Bright compared this account with the linked original and supporting sources and kept reported, budgeted, projected, and observed claims distinct. Bright did not independently audit the underlying records.
Maturity describes the tested or operational setting. Confidence describes support for the particular claim; one does not determine the other.
Revision & correction history
No corrections recorded.
Keep exploring
Explore the shared question in another setting. These connections do not imply replication.