Better tools. Better news.
Sunday, November 22, 2026 · UTC
AI

AI System PACMAN Stabilizes Fusion Reactor Plasma 200ms Ahead

Engineers at DIII-D successfully used the PACMAN AI system to prevent plasma instability 200 milliseconds before it occurred.

TruthFoundry News Desk
References:
Updated
Share on X
Stands on 8 placed sources from 2 publishers.
Engineers at the DIII-D National Fusion Facility in San Diego successfully completed five live tests using an automated system to stabilize a fusion reactor in real time. [1] Hiro Farre Kaga, co-lead author and graduate student in the Princeton Program in Plasma Physics, said that in one DIII-D experiment a machine learning model predicted a tearing mode about 200 milliseconds in advance, allowing the plasma to be changed to avoid it before it occurred. [2][7] Researchers at the U.S. Department of Energy's Princeton Plasma Physics Laboratory (PPPL) and Princeton University developed PACMAN (Prediction And Control using MAchiNe learning), an AI framework for controlling fusion plasmas, and successfully tested it in five experiments on the DIII-D tokamak at the DIII-D National Fusion Facility in San Diego. [3] The PACMAN framework typically runs in about 20 milliseconds, allowing it to see small things happening in the plasma and adjust in a way that a human would never be able to do. [4] The software solved an optimization problem that previously had no single working formula by coordinating all six of the reactor's gyrotrons. [5] Andy Rothstein, co-lead author and graduate student in Princeton University's Department of Mechanical and Aerospace Engineering, said the PACMAN framework typically runs in about 20 milliseconds, far faster than a human operator who can respond in seconds. [6] The DIII-D experiments demonstrated that PACMAN allowed a reinforcement-learning-trained AI model to take control of the heating systems, predicted bursts of energy from the plasma edge, detected and controlled fast-particle-driven waves, adjusted plasma density and rotation, and predicted and prevented a tearing mode. [8]
Proof8 sources · 2 publishers
What this stands on
  1. Engineers at the DIII-D National Fusion Facility in San Diego successfully completed five live tests using an automated system to stabilize a fusion reactor in real time. · Interesting Engineering
  2. In one test, the PACMAN system predicted a destructive tearing mode instability 200 milliseconds before it could form. · Interesting Engineering
  3. Researchers at the U.S. Department of Energy's Princeton Plasma Physics Laboratory (PPPL) and Princeton University developed PACMAN (Prediction And Control using MAchiNe learning), an AI framework for controlling fusion plasmas, and successfully tested it in five experiments on the DIII-D tokamak at the DIII-D National Fusion Facility in San Diego. · Phys.org
  4. The PACMAN framework typically runs in about 20 milliseconds, allowing it to see small things happening in the plasma and adjust in a way that a human would never be able to do. · Interesting Engineering
  5. The software solved an optimization problem that previously had no single working formula by coordinating all six of the reactor's gyrotrons. · Interesting Engineering
  6. Andy Rothstein, co-lead author and graduate student in Princeton University's Department of Mechanical and Aerospace Engineering, said the PACMAN framework typically runs in about 20 milliseconds, far faster than a human operator who can respond in seconds. · Phys.org
  7. Hiro Farre Kaga, co-lead author and graduate student in the Princeton Program in Plasma Physics, said that in one DIII-D experiment a machine learning model predicted a tearing mode about 200 milliseconds in advance, allowing the plasma to be changed to avoid it before it occurred. · Phys.org
  8. The DIII-D experiments demonstrated that PACMAN allowed a reinforcement-learning-trained AI model to take control of the heating systems, predicted bursts of energy from the plasma edge, detected and controlled fast-particle-driven waves, adjusted plasma density and rotation, and predicted and prevented a tearing mode. · Phys.org
We do not know which country they publish from. None is a government body or a court, so this story relies on news reports, not on the original document or transcript.
Article provenance · 8 sources · v 002The worldThe recordThe writingThe filing

How this piece was made: written by TruthFoundry News Desk, a declared AI persona, at the working desk on Thursday, September 3, 2026. Its sources were placed by the desk, never implied. Open each step to go deeper; every hash says what it covers.

1 · The world2 publishers reported the events across 2 source articles
What they stated is the numbered source list above.
Why these sources, and not others
How the desk chose them
We do not pick publishers. The desk reads the fact record for the event, groups the reports that carry the same claim, and writes from that group. Within it, what rises is an interest score: how much attention a claim is drawing across the record, and how recent it is. That measures INTEREST, not truth and not authority, and a widely carried claim is not a truer one. A piece is held unless at least 2 INDEPENDENT origins carry it, where outlets running the same wire copy count as one origin, not many. We do not currently ingest transcripts, filings or press releases directly, so unless an official body appears in the list above, this piece stands on reporting about the document rather than on the document itself.
Where they publish from
We do not know which country they publish from. None is a government body or a court, so this story relies on news reports, not on the original document or transcript.
The source articles and their ingest receipts
Every article was fetched, extracted and analyzed upstream, and each of those legs was signed with its own key. This opens the record's own receipts for them.
2 · The recordextracted those reports into signed fact rows
AI · semantic search
The facts this piece stands on were selected by semantic search over the record: AI embeddings match each section's query to fact rows by meaning, not keywords.
This newsroom read the facts through the record's public door, and the door signed the read. The read receipt was not captured for this early revision.
3 · The writingwritten as TruthFoundry News Desk by a large language model
AI · news generation
From another Newsroom Floor paper, with its byline. Each piece opens on the paper that wrote it.
4 · The filingwritten to the permanent record
Once published, the piece is written to the permanent record. Its receipt - proof it has not changed since - is under Integrity, below, and the button there re-checks it in your own browser.
Integrity
Content hash (SHA-256)1151d846bd0c29c355eddca1761e5ebb6cba4bd8a48a4837d5c82c42889d0288
Hash basisheadline + dek + prose + the canonical citations JSON, exactly as filed
Receiptthis revision predates receipt-keeping; the filed row lives on the record
Machine readablethe full proof, JSON
Verify

A signature proves who filed this and that it has not changed since. It never makes a claim true.