Skip to content

ONR leads expert discussion on AI in nuclear industry

An expert panel selected from leaders within the nuclear sector recently met to discuss opportunities for the regulation of Artificial Intelligence (AI) in the industry.

Jointly organised by the Office for Nuclear Regulation (ONR) and the Advanced Nuclear Skills and Innovation Campus (ANSIC), the seminar panel was hosted by RACE (Remote Applications in Challenging Environments), part of the United Kingdom Atomic Energy Authority (UKAEA), at the Culham Science Centre, near Oxford.

ANSIC was a Department for Business, Energy and Industrial Strategy (BEIS) funded initiative led by National Nuclear Laboratory.

The expert panel comprised representatives from a broad range of organisations including nuclear licensees, regulators from other sectors, academics and national laboratories.

These included EDF Energy, Rolls-Royce SMR, Sellafield Ltd, UKAEA and the Universities of Bristol, Manchester and Oxford.

This AI event aligned to ONR's ongoing commitment to embrace innovation, new approaches and technologies in how we regulate - sharing best practice case studies and encouraging dialogue, in conjunction with the ONR Strategy 2020-25.

Part of the debate focused on reviewing proposals from panel members for potential applications of AI that could be challenging to permission, for taking into ONR’s new regulatory sandbox.

Sandboxing enables innovators to test and trial new solutions in a safe environment which enables the safe and secure adoption of innovation to the benefit of society.

At Culham, it was agreed to take two proposals forward to the next stage of the sandbox process.

David Smeatham, Innovation Lead at ONR, said: "We are committed to considering new ideas and innovative approaches in the nuclear industry, while still delivering compliance with the law and ensuring that the high standards of safety and security are met.”

Gary Bolton, Panel Chairman, National Nuclear Laboratory added: "It is clear that there are many opportunities for the application of AI in nuclear.

"However, there is currently no regulatory approach for the use of AI in the nuclear industry so it was great to see a wide variety of stakeholders come together and share information to potentially enable regulation of AI in the nuclear sector.”

This latest panel event followed a workshop at Bristol University earlier this summer where speakers, including David Smeatham, Innovation Lead at ONR, examined how the nuclear industry could use data science to gather and store information to support future safety cases.

AI could potentially be used in the nuclear industry to simulate behaviour of reactors, and inform reactor design, performance, safety and operation.

A similar panel took place last March, involving ONR and NNL, which helped to set out how we are open and responsive to facilitating the implementation of innovative solutions within our regulation.

Read more about ONR's approach to regulating innovation.