Breaking New Ground In Nuclear Innovation

The first International RegLab brought together the NEA, Ennuvo, IAEA, EPRI, and national regulators to examine how AI could support real‑time monitoring of large volumes of sensor data to detect deviations from normal plant behaviour. This collaborative “safe space” enabled experts to explore innovation while maintaining the nuclear sector’s high safety expectations.

The collaborative work revealed that the potential for enhancing safety and reducing costs is significant. The RegLab considered the following three areas:

• For high consequence tasks, we cannot rely on AI "explaining" its logic alone. We must maintain "defence-in-depth"—using traditional physics-based models and engineering backups to ensure the plant remains protected even if the AI falters.

• We need to shift our focus from the use of large quantities of data to ensuring data quality and governance. AI is only as reliable as the information used to train it, and establishing industry-wide standards for this data is needed.

• As we adopt these tools, we must proactively address the "skills gap" to ensure our workforce remains empowered to oversee and challenge AI-driven decisions.

Read the full Report here.