Government proposals for UK AI regulation
The Government today released a policy statement setting out proposals for AI regulation in the UK. The paper places a strong emphasis on establishing a framework which encourages AI innovation, setting out the ambition to be context-specific, pro-innovation and risk-based, coherent and proportionate and adaptable. In techUK’s paper Governance for an AI future, we highlighted the essential need for any regulatory measures to be risk-based and to avoid contradiction or replication of existing regulation, and therefore welcome the Government’s recognition of the importance of this.
Today's statement reveals that the Government is not planning to establish a central AI regulator, and is instead intending to provide sector regulators with cross-sectoral principles for them to apply to AI developed and used within their remits. The proposed principles are:
- Ensure that AI is used safely
- Ensure that AI is technically secure and functions as designed
- Make sure that AI is appropriately transparent and explainable
- Consider fairness
- Identify a legal person to be responsible for AI
- Clarify routes to redress or contestability
When it comes to defining the scope of AI regulation, the paper argues for an approach which sets out core characteristics of AI technologies, but does not settle on a fixed definition. The intention would be to target regulation at the use of AI rather than the technology itself, and to allow individual regulators to set out and evolve more detailed definitions in line with their specific domains or sectors. The Government has identified two core characteristics of AI which they believe prompt the need for a regulatory framework for AI; its ‘adaptiveness’ and its ‘autonomy’. The regulatory principles summarised above are intended to address resulting challenges relating to explaining the intent or logic of an AI application, and assigning responsibility for action resulting from it.
The government has now launched a call for views which will be open for 10 weeks, after which the white paper announced in the National AI Strategy will be finalised and published by the end of the year. techUK is hosting our first meeting with members to discuss our response on Wednesday 20 July, 13.30-14.30. If you would like to join this, please get in touch with [email protected].
Sue Daley, Director of Technology and Innovation at techUK commented on the proposals:
techUK welcomes the Government's intention to pursue an innovation-friendly AI governance regime. We are pleased to see that the policy position paper published today highlights the need to account for the specific contexts in which AI is being used and that duplication or contradiction between regulatory regimes must be avoided. These are both key issues highlighted in techUK’s paper Governance for an AI future published earlier this year.
The UK is well placed to be a leader on AI governance given its world-class research institutions, regulators and civil society organisations. We encourage the Government to prioritise the completion of the white paper setting out its approach in full, as regulatory certainty is a high priority for innovators and companies are already assessing and adapting to other international proposals and initiatives on AI governance.
techUK looks forward to working with our members and the Government to inform a timely outcome, which can bring about greater confidence in AI, encouraging greater adoption and thereby fuel the economy and improve standards of living across the country.
The Government also published an AI Action Plan today, which provides a useful overview of progress made since the publication of the National AI Strategy last year. You can find the Action Plan here.