Charting a course on regulating AI

Back in June last year, ADS made a submission to the Government setting out our sectors’ views on the hot topic of the regulation of artificial intelligence (AI) systems. In our response to the Government’s White Paper (A pro-innovation approach to AI regulation), we supported the UK’s light-touch, context-sensitive approach but called for a beefed-up central unit and adequate resourcing for the impacted regulators, amongst other things. Towards the end of 2023 the Government then held a high-profile summit on frontier AI and AI safety, which culminated in the Bletchley Declaration and the launch of an AI Safety Institute.

Following on from this, the Government has now issued its response to last summer’s consultation, providing an important update on how the UK plans to regulate AI. The key things to know are below:

• A number of regulators have already set out their strategic approach to AI (notably, the Civil Aviation Authority released their strategy for AI at the same time as this response). However, for those outstanding (including the Competition and Markets Authority) they have been asked to outline their plan by 30 April 2024.
• £10m has been announced to prepare and upskill regulators, which was a key ask from ADS and is a much-needed measure given the challenges AI poses both from a capacity and expertise perspective.
• The UK will maintain its sector-based approach to regulating AI and will not introduce any immediate AI-related legislation. However, the Government has now made clear it is minded to introduce binding requirements in future on the developers of frontier AI systems.
• Nearly £90m has been allocated to support the launch of nine new research hubs across the UK (with areas of focus including chemistry and mathematics) and a partnership with the US on responsible AI.
• A central, formal regulator coordination structure will be introduced, in keeping with ADS’ proposals for a central unit last year. This will be driven forward by a new steering committee that will consider the UK’s AI risk register.
• A pilot advisory service for helping business to launch AI and digital innovations will be established, with support from several expert regulators and backed by over £2m of funding.

Navigating through the risks and opportunities that AI presents is a common challenge for both Government and industry. Given the rapid improvements in AI capabilities it also calls for an agile and pragmatic approach to ensure the UK can reap the rewards of this technological revolution. As such, these can only be welcomed as a very positive set of announcements by the UK, while accepting that the Government must continue to engage closely with industry, including our sectors, on the development and deployment of AI for our collective benefit.