A New UK Labour Government: A Fresh Approach to AI Regulation

As the dust settles from the Labour Party’s landslide victory in the UK’s General Election on 5 July 2024, and with the UK AI market projected to surpass $1 trillion by 2035¹, we review the future of artificial intelligence (“AI”) regulation in the UK under the new UK Government.

The UK’s Approach to Regulating AI

The previous Conservative Government championed a ‘pro-innovation’ approach to AI regulation².  As part of this strategy, the UK developed a non-binding, cross-sector, principles-based framework to enable existing regulators such as the Information Commissioner’s Office, Ofcom and the Financial Conduct Authority to apply bespoke measures within their respective fields of data protection, telecommunications and finance. While this approach anticipated that there may be a need for targeted legislative interventions in the future, specifically for General Purpose AI systems, it prioritised remaining agile as new technologies emerged.

Sir Keir Starmer, the new UK Prime Minister, previously suggested that a Labour government would move away from the Conservative government’s laissez-faire, pro-innovation strategy. Instead, Labour intends to introduce stronger regulation of AI, albeit in targeted areas. Starmer has publicly emphasised the need for an overarching regulatory framework and has expressed concerns about the potential risks and impacts of AI, while also acknowledging its transformative potential for society. The Labour Party’s manifesto specifically mentions implementing binding regulations on the “handful of companies developing the most powerful AI models” and prohibiting the creation of sexually explicit deepfakes.

The UK’s approach to regulating AI has stood in stark contrast to the EU’s AI-specific legislation, described by the European Parliament as ‘the world’s first comprehensive AI law’, which, after several delays, is expected to come into force later this summer. The Labour Party’s manifesto does not appear to radically diverge from the previous government’s light-touch approach and suggests that any forthcoming legislation will be more narrowly focused around “powerful AI models”. However, there are no additional details on what will constitute a “powerful AI model”³. UK-based organisations with operations in the EU or those deploying AI systems within the EU would still likely fall under the jurisdiction of the EU AI Act, therefore requiring UK organisations to keep abreast of legislative changes and any potential future alignments between the UK and EU in this area (see our OnPoint for an overview of the EU AI Act).

Initiatives Under the New Labour Government

The Labour Party’s manifesto outlines several key initiatives related to AI under its ‘Kickstart economic growth’ section, which states that “markets must be shaped, not merely served”⁴:

• Regulatory Innovation Office and Developing Ethical AI - Labour aims to address the regulatory gap concerning AI by establishing a ‘Regulatory Innovation Office’. It is proposed that the new office will consolidate government functions, streamline approval processes for innovative products and services and manage cross-sectorial issues. It will also set targets for technology regulators, monitor their decision-making speed against international benchmarks and guide them according to Labour’s industrial strategy. This office is intended to introduce binding regulations on companies developing powerful AI models to ensure their safe development and will ban the creation of sexually explicit deepfakes. However, specific details on these binding regulations are not yet available.
• The Regulatory Innovation Office will not be a new AI regulator, therefore, existing regulators will still address AI within their respective spheres. It also remains to be seen how the Regulatory Innovation Office will materially differ from the Conservative Government’s proposals to deliver ‘central functions to support the framework’⁵, or the ‘AI Safety Institute’, which the Conservative Government established at the beginning of 2024 (as the first state-backed organisation focused on advanced AI safety for public interest⁶).
  
• Support for Data Centres - To support the growth of AI, the Labour government plans to remove planning barriers for new data centres by designating them as Nationally Significant Infrastructure Projects. This reclassification would allow these projects to circumvent local opposition and to consequently speed up their approval process.
• Creation of a National Data Library - The National Data Library initiative is a component of the Labour Party’s broader national industrial strategy. It aims to consolidate existing research programmes and to help deliver data-driven public services “whilst maintaining strong safeguards and ensuring all of the public benefit”.
• Long-term R&D Funding - Labour committed to scrapping short funding cycles for key R&D institutions in favour of ten-year budgets that should allow for meaningful industry partnerships. The government will collaborate with industry to support spinouts and start-ups by providing the necessary financing for their growth. This initiative aims to simplify the procurement process and foster innovation.

Future Outlook

With the new Labour Government in power, the technology sector can likely expect a shift towards more proactive and structured regulatory measures. While there have been indications of an intention to implement stricter regulations around AI, there has been no proposal for a general AI regulation. Any new legislation is expected to be more narrowly focused than the approach taken by the EU. The Labour Government’s manifesto suggested that the UK will maintain a relatively light-touch regulatory approach to AI. However, the Labour Party ran a cautious election campaign and, having won, it proposals may become bolder. In addition, its plans to re-build a closer trading relationship with the EU may lead to greater alignment with EU regulation. Industry stakeholders should therefore remain informed to effectively navigate what is likely to be  an evolving regulatory landscape.