The BaseResearcharXiv
Deploying a Control-Logic LLM Assistant in Air-Gapped Power Plants: Lessons in Source Authority and Bounded Retrieval
According to the abstract excerpt, Fu and Gabriel report lessons from deploying an on-premises assistant for control logic at two gas power plants. Two prototypes showed where generative models added avoidable uncertainty and cost, and simpler designs replaced them.
Why it matters
The same thing, for different readers.
What it means for you
Authors working at two gas power plants report that a plain structured export beat an AI model at reading control diagrams.
For researchers
Fu and Gabriel report extraction at 0.51 connection precision taking 3.1 hours on 4xH200 GPUs, and a 20x token and 3x latency reduction from explicit plans on 40 template questions.
How the finding was reached
In our words, from the source. Weigh the finding by how it was made.
- What was done
- Two prototypes in on-premises operator assistance at two gas power plants: a vision-language model reading a 6,979-page document of printouts, and an LLM tool-calling agent tested on 40 template-generated questions.
- What it found
- Per the authors, extraction reached 0.51 connection precision in 3.1 hours on 4xH200 GPUs, and explicit execution plans cut mean token use by 20x and median latency by 3x.
- What it does not show
- A controlled benchmark of template-generated questions at two sites; the excerpt is cut off.
- Status
- A preprint: not yet peer reviewed
The record
- Questions
- Are AI companies putting stronger safeguards into practice? · How is AI used in attacks, defence and war?
- People
- Yiwei Fu (Wrote it) · Nicholas Gabriel (Wrote it)
- What we read
- We have read only the description the publisher gives. No quotations until someone reads the source itself.
- Found by
- The Radar, run of November 17, 2026, via
arxiv-ai-safety-query(tier 1: the source’s own channel or an official record) - Cite this
- arXiv, “Deploying a Control-Logic LLM Assistant in Air-Gapped Power Plants: Lessons in Source Authority and Bounded Retrieval”, November 16, 2026. Entry in the Ctrl AI Base, https://ctrlai.com/base/arxiv-2026-11-16-llm-assistant-air-gapped-power-plants
Our notesLight lane: written from the triage excerpt only. Tangential to AI control; kept as an industrial deployment lesson.