It’s often claimed that there are no votes in AI, but with the opportunities presented by new technology, this could be the most exciting and expansive time to be in government, says Benedict Macon-Cooney
Too much time and energy has gone into the Prime Minister’s AI strategy when “there isn’t a single vote in it,” was the criticism in this weekend’s Sunday Times. There may be some truth in its electoral attractiveness as fears about its impact on jobs and privacy dominate public discourse – although AI’s biggest and fastest impact will be felt most in the areas voters care most about (ending waitlists, keeping people healthy and giving ever child a quality education). But there are signs the UK is taking the right approach. It’s just not ambitious or fast enough to reimagine the state.
A large part of the UK’s success in producing world-leading science and technology has been historical but there are reasons to think we’re catching up with ourselves. Home to leading thinkers such as Alan Turing and Geoff Hinton, who shaped the course of AI’s development, more recent signs of hope include Demis Hassabis and DeepMind, who have laid a foundation for our ability to get anywhere close to the strength of Silicon Valley. This fact was underscored by his co-founder, Mustafa Suleyman, who is now leading Microsoft’s AI efforts, yesterday announcing that the company would launch a new AI hub in the UK. It comes after a raft of companies have announced UK operations over the last 12 months. Generative AI leader Aiden Gomez made the UK a centre of operations for his company, Cohere, while those who opened up this explosion, OpenAI, have done the same.
In biotech, the drug discovery company Recursion recently joined DeepMind-affiliated Isomorphic Labs in Kings Cross in London, developing a cluster that could hopefully rival Boston in the US, where many of the big beasts are pushing the boundaries of biology. Added to this mix are investors such as Andreessen Horowitz and Flagship who add both significant capital and expertise to the sector here.
Home-grown lights, such as Faculty for AI adoption, Helsing for defence, and Exscientia for biotech, mean that our industry is in good health. If one were to bet on the UK’s first trillion-pound company, it would take a brave person to look past those in AI. And as we look to arrest some of the UK’s stagnant productivity, ensuring that both the next wave of companies continues to grow and scale here, while wider British PLC adopts emerging technology, will be critical.
The conditions that we set are important for the companies to launch and scale and the Prime Minister has been right to maintain a pro-innovation stance for AI in health, climate and education, while leading the way on the Safety Institute – an idea that has also now been taken up the United States, Singapore and Canada in the past week.
But a wider prize is also available. As a general-purpose technology AI will impact everything. From where we are, the pace of change is only set to increase. More capable models. Cheaper and wider diffusion. More and more application into everyday life.
Historically, government has been slow to adapt to technological change, but in the UK we have seen some promising signs of learning from the past.
The new Incubator for AI in the centre of government has started to find returns beyond expectations: having set a return on investment target of 3.55, some of the initial tools have been around 200 times. Or as the recent NAO report estimated, one third of civil services tasks can be automated via AI.
Scaled up, these kinds of changes will allow government to be far more strategic in how it deploys its resources, driving down the cost of government while improving outcomes. It will also free up our public servants to spend more time serving the people.
The signs may only be small at the moment, but through AI-enabled health, AI tutors for education, and to help solve crime, this technology can be transformative. Added to this are the general processes of government such as planning, which can all add up to build a new model for government in the 21st century.
But we won’t get there if we allow politics to get locked in a battle between different forms of conservatism. If we are consumed by old debates on tax and spend, the size of the state, or individual versus state control, we not only set our sights too low, but also ignore the major change and opportunity that AI presents.
Working closely with private sector will be critical for this. But at a time when we worry about the state of the public finances, a consistently weak economy and crumbling public services, this moment might seem like the most limiting in living memory to be in government. In fact, if you consider the opportunities now presented by technology, it might yet be the most exciting and expansive. Harnessed in the right way, it should be a vote winner.
Benedict Macon-Cooney is Chief Policy Strategist here at the Tony Blair Institute