30 September 2025
In a recent study conducted under the direction of the Operations and Safety Department of the International Union of Railways UIC, we explored the issue of Artificial Intelligence and safety/certification.
We tested an original approach, inspired by MIT:
- entrusting AI with solving the complex problem
- provided that all its answers are validated by an independent safety controller
This method has been successfully applied to a concrete and comprehensive case study: detecting that a track is free of obstacles in the railway sector.
This experiment has given rise to a methodology and a specific recommendation, which will soon be published by the UIC.
This breakthrough shows that it is possible to harness the power of Artificial Intelligence without compromising on security.