Question to the Cabinet Office:

To ask the Minister for the Cabinet Office, with reference to the Government's publication A pro-innovation approach to AI regulation, published on 29 March 2023, how much and what proportion of the budget of each regulator in their Department was spent on regulation of artificial intelligence in the latest period for which information is available; how many staff in each regulator worked (a) wholly and (b) partly on those issues in the latest period for which information is available; and whether those regulators plan to increase resources for their work on artificial intelligence.

The AI White Paper emphasised the importance of ensuring that UK regulators and public bodies have the capacity, expertise, and capabilities to implement government’s pro-innovation approach whilst recognising and understanding the risks. This is particularly true for those regulators for which AI falls squarely within their regulatory remit, but also applies to a much wider range of public and regulatory bodies considering the implications AI has across the economy.

The Cabinet Office ALBs include 3 bodies that are either formally or informally classified as regulators:

1. The Civil Service Commission: The CSC are often referred to as a ‘regulator’ of recruitment into the Civil Service but undertake assurance/compliance work in respect of civil service recruitment only and are out of scope.

2. The UK Statistics Authority includes the Office of Statistics Regulation who set the statutory Code of Practice for Statistics and assess compliance with the Code. The Office for Statistics Regulation (OSR) utilises a regulator on a 0.33 FTE basis for their work providing guidance on how the principles in the Code of Practice for Statistics can help in designing, developing and using models to improve their trustworthiness, quality and value. Their guidance covers both traditional statistical techniques, such as linear regressions, and newer techniques, like machine learning, when they are used to create outputs that inform decision making or public policy. The OSR also regularly engages with other regulators on this topic through the attendance of workshops and working groups to ensure they remain aware and responsive to any developments in this space. The OSR does not plan to increase the time allocated to these issues. The promotion of the Code of Practice for Statistics into new areas of statistical analysis and data use has been a long-standing area of focus for the next year, and will continue to be, including in relation to AI (e.g. large language models).

3. The Equalities and Human Rights Commission: the Equalities Hub have provided a separate response on this PQ (188555) via the Rt Hon Kemi Badenoch MP, Minister for Equalities.

As part of the AI regulation White Paper consultation, we are engaging closely with regulators across the wider landscape and their sponsoring government departments to understand the organisational capacity they need to regulate AI effectively, across technical, regulatory, and market-specific expertise. This will inform our work to develop policy options with a view to addressing any gaps that emerge.