Artificial intelligence-related applications for railway security

In an increasingly strong security context, railway companies worldwide are exploring the growing potential of Artificial Intelligence. In the past years, those technologies demonstrated that they can represent a key factor and a major asset to increase efficiency by tackling many kind of issues, even in the security domain. Especially, the emergence of Machine and Deep Learning, followed by the Large Language Model solutions, further accelerates this deep transformation in the positioning of technologies within the chains of command of any companies.

In this regard, applications of AI systems may support security experts within railway companies at each step of the security cycle in many different activities towards the protection of assets and persons, potentially including real time surveillance and reporting, access control, early detection of weapons, violence and other potentially dangerous behaviours, incident response and post-event analysis.

All the above-listed use cases are highly critical, implying personal data processing and ethics considerations, and many railway companies are now willing to start testing or implementing some of those, facing sometime difficulties related to the legal framework that is also facing the regulation of AI technologies challenge.