Skip to content
Search
paperFebruary 2026Unreviewed

Osprey: Production-ready agentic AI for safety-critical control systems

Thorsten Hellert, João Montenegro, Antonin Sulc

APL Machine Learning

View resource

Abstract

Operating large-scale scientific facilities requires coordinating diverse subsystems, translating operator intent into precise hardware actions, and maintaining strict safety oversight. Language model-driven agents offer a natural interface for these tasks, but most existing approaches are not yet reliable or safe enough for production use. In this paper, we introduce Osprey, a framework for using agentic AI in large, safety-critical facility operations. Osprey is built around the needs of contr

Categories

Agentic Threats

Cite

@article{hellert2026osprey,
  title = {{Osprey: Production-ready agentic AI for safety-critical control systems}},
  author = {Thorsten Hellert and João Montenegro and Antonin Sulc},
  year = {2026},
  month = feb,
  journal = {APL Machine Learning},
  doi = {10.1063/5.0306302},
  url = {https://doi.org/10.1063/5.0306302}
}