What To Expect for AI From the UK’s New Labour Government
The UK’s new Labour government has inherited a pro-innovation regulatory framework for AI, one that heavily leverages sector-specific expertise to govern this technology. While the framework has been well-received globally, Labour’s manifesto hinted at stronger policy interventions for AI, as well as a desire to use AI as an enabler for key ambitions like data-driven public services. Following the King’s Speech two weeks ago, which outlined the new government’s legislative agenda, three key issues are likely to dominate UK AI policy discussions: AI legislation, boosting the UK’s AI ecosystem, and unlocking AI-driven productivity in the public sector. Crucially, the thread tying these three issues together is Labour’s desire for economic growth, which the UK technology sector should take as a positive step in the long road to homegrown AI innovation.
UK AI Legislation on the Horizon
In contrast to what most people were expecting following the King’s Speech, Labour omitted an explicit AI bill from its first package of proposed laws. Instead, the Speech mentioned “appropriate legislation to place requirements on those working to develop the most powerful artificial intelligence models.” Whilst it is unclear specifically what the legislation will entail, taking a stronger stance on frontier model AI development companies invariably means targeting those larger tech companies based outside of the UK and in particular, based in the United States, which may impact transatlantic relations. For example, UK legislators could require companies to implement the voluntary commitments made at the latest mini-AI Safety Summit in Seoul (which would deter future voluntary commitments) or create formal requirements for companies to work more proactively with the UK AI Safety Institute. In any event, UK policymakers should look to the EU as an example of how not to regulate AI, not only to re-establish independence from the EU, but to afford themselves the opportunity to foster a pro-AI innovation regulatory landscape that fuels Labour’s economic growth ambitions.
Whether intentional or not, the announcement of AI legislation followed very shortly after the announcement of the Employment Rights Bill. Workers’ rights are at the forefront of Labour’s agenda, and it would therefore be unsurprising if proposals emerged from the party around AI in the workplace. But Labour should proceed cautiously. Whilst fears around AI hurting jobs are certainly rife, with a recent report highlighting that 66 percent of respondents in the UK were concerned about AI-driven job displacement, AI will not drive workers to a jobless future. Similarly, existing employment legislation and anti-discrimination laws already capture unfair practices such as discriminatory hiring and firing, even if it involves AI, so additional legislation is unnecessary.
Finally, though no mention as of yet around the establishment of a new Regulatory Innovation Office (RIO), touted by Secretary of State Peter Kyle MP, a pro-innovation body would be a sensible counterweight to potentially over burdensome regulation under a Labour government. Its mandate should include evaluating regulation through the lens of innovation (which would support the potential for future AI legislation), advocating for pro-innovation measures (which would fuel the desire for broader economic growth), and helping UK regulators responsible for sector-specific implementation (which would reinforce the sector-specific approach to AI regulation).
However, broader regulatory capacity will remain an issue across sectors as regulators continue to come to grips with AI-related risks and opportunities, and it is unclear how Labour will respond to this growing expertise gap. The Department for Science, Innovation and Technology (DSIT) should carry out a fact-finding exercise to establish where regulators are currently struggling, such as by working with the Digital Regulation Cooperation Forum, a forum that brings together key UK regulators including the Competition and Markets Authority and the Information Commissioner’s Office. These findings could then form part of the basis of a wider mandate for the RIO.
Boosting the UK’s AI Ecosystem
The Labour government has also promised in their manifesto a more concerted effort to attract late-stage capital investment into the UK for deep tech, including AI, by expanding the risk appetite of pension funds and relying on public-private relationships to leverage foreign investment. Whilst greater investment is needed to grow the UK economy, this move is unlikely to make much of a dent, particularly as the Conservative government has already explored this option, indicating the need for a stronger industrial policy strategy.
That being said, Labour seems keen to deploy AI more widely across the UK, evidenced by the manifesto's promise to reform planning decisions that allow government ministers to override local council decisions that create barriers to building data centres. Moreover, the promise of a new Digital Information and Smart Data Bill (DISD), a possible resurgence of the Conservatives’ Digital Information and Data Protection Bill, could complete what the previous government was unable to finish, simplifying compliance with the General Data Protection Regulation (GDPR) which would offer greater clarity both for industry and government where civil servants are fearful of committing anti-GDPR practices.
Finally, the announcement in the manifesto of a new National Data Library could be a good starting point for capitalising on the wealth of public data on which the government currently sits. However, then Secretary of State Michelle Donelan made similar announcements for startups and scale-ups, providing access to government-held data to help generate data-driven solutions. This formed part of a broader National Data Strategy which unfortunately lacked centralised leadership to drive it forward. Should Labour want to learn from the mistakes of the last government, it should think more broadly about UK data infrastructure, and how to promote data sharing between bodies that can be funneled into a national AI ecosystem. The new DISD Bill could achieve this, as the government itself has said the bill would, amongst other things, set up a Smart Data scheme for secure sharing and reform the Digital Economy Act to “help the Government share data about businesses that use public services.”
AI Productivity Gains to Compensate for Public Sector Cuts
One issue remains clear, however. UK productivity is lagging, and Labour has shown a desire to rely on AI to rectify this. The previous Conservative government initiated an AI-driven productivity agenda within public services when then chancellor Jeremy Hunt announced in his 2024 Spring budget the landmark Public Sector Productivity Plan, using AI to “cut down and potentially cut in half form filling by doctors” and improving MRI and CT scans. Labour has made similar commitments to introduce AI into healthcare, and increased reliance on AI by the public sector could afford the government the justification needed to make public sector cuts for further cost-saving measures. With promises of productivity gains of close to £1 billion a year for bodies like the Department for Work and Pensions (DWP), AI is likely to feature heavily within a UK public sector strategy. This would throw more pressure on departments like DSIT to roll out AI solutions quickly, which is perhaps why one of the first announcements Peter Kyle made as Secretary of State was to merge DSIT with other central government organisations like the Government Digital Service, the Central Digital and Data Office, and the Incubator for AI. Yet, even if DSIT develops AI solutions, Labour will no doubt encounter challenges with integrating these into legacy systems, and transitioning and preparing public sector workers, and the UK public, for an AI-powered service. The government should therefore explore national initiatives that support this transition, often dubbed in the technology industry as “change management.” Such initiatives could involve a combination of upskilling programmes for civil servants, dedicated AI champions and change agents deployed to local public services across the UK, and staggered pilot initiatives that go beyond traditional cities like London.
Top-level concerns for the UK around global competitiveness mean AI will feature as one of many solutions to tackle issues like slowed productivity. In contrast to the previous government however, where much of the focus was kept on AI, this new post-election era may take a more muted approach to AI governance, viewing the technology as an enabler for much-needed change, and underpinned by the overall desire for economic growth.
Image credits: UK Government
