Discussion Series: Artificial Intelligence in Nuclear Waste Management

Participants of the first AI panel organised by BASE Source: BASE

What are the potentials and limitations of using artificial intelligence (AI) in nuclear waste management? Selected experts met on 1 December to get to the bottom of this question, and develop initial answers. The invitation had been extended by the Federal Office for the Safety of Nuclear Waste Management (BASE).

AI is currently being discussed in many areas of society: customer service, journalism and science. Intelligent programmes can take over human tasks everywhere - sometimes with astonishing results. AI methods are also gaining importance in the search for a final repository for high-level radioactive waste.

This applies, in particular, to technology and the natural sciences, where AI applications are to reduce uncertainties, for example in the computer-assisted modelling of phenomena and processes. With regard to the social and political sciences, AI may become relevant in the future for the systematisation of data material or in processes for public participation.

Can AI make a repository safer?

But what does that mean in concrete terms? Can these technologies improve decisions regarding the repository search? Can their use actually lead to greater safety and reduce risks?

Such questions have rarely been asked in the context of the repository search so far, and they provided the impetus for the BASE discussion series. The contributions at the opening event were manifold. They addressed the following issues:

• The prerequisites for the use of AI methods in nuclear waste management,
• the potential of AI in dealing with geo-data,
• experiences from other environmental policy application fields with AI,
• findings from the AI-supported evaluation of participation processes, which might also become relevant in the search for a repository, as well as
• ethical and legal framework conditions that already define the use of AI today, and that are currently being negotiated at EU level with a view to the 'Artificial Intelligence Act'.

Quality of data is key

The discussions made it clear that AI methods certainly bear potential for the disposal of radioactive waste. However, limitations became apparent, particularly with regard to the existing data material and the availability of data. If AI is to be used for image recognition, for example, sufficient image material must first be available to train the AI method applied. In addition, corresponding IT infrastructures as well as human resources are needed wherever AI is used.

Furthermore, it is unclear how solid the results of AI methods are, and whether AI applications should not rather function as 'control instances' that reduce errors and uncertainties.

Actors involved in the search for a final repository also face questions of their own: What would happen, for example, if the Bundesgesellschaft für Endlagerung (BGE) actually used AI in its search for the most suitable site for a final repository? As a supervisory and licensing authority in the repository search process, BASE would have to identify potential regulatory needs.

The panel kicked off a series of AI-related discussions in the context of nuclear waste disposal, which will be hosted by BASE.