What exactly is the government’s AI strategy?
The government has so far scrapped £1.3bn in AI and tech funding – including the much-lauded Edinburgh supercomputer, announced 98 “new” AI projects and established an “AI Action Plan Unit.” So, what exactly is the government’s AI strategy?
Why scrap the Edinburgh supercomputer if economic growth is the goal?
It was a busy and confusing Summer for those trying to make sense of the new government’s approach to AI. On the 26th July it announced the hiring of Matt Clifford, tech entrepreneur and chair of Aria, to deliver an AI Opportunities Action Plan for the UK. A positive move. Then, barely a few weeks later it scrapped £1.3bn in funding for major tech and AI projects – including the UK’s first exascale supercomputer being built at Edinburgh University. A further £500m for AI Research Resource, which was assigned to computing power for AI, was also scrapped.
As if that wasn’t confusing enough, the Department for Science, Innovation and Technology (DSIT) then announced – barely a week later – new funding for 98 specific AI projects.
Was this latter announcement a case of fallout management played out in real-time, or does it give us an indication of the new government’s approach to AI, underpinned by its broader fiscal commitment to balance the books? Albeit delivered with the communications finesse of The Thick of It’s Malcolm Tucker…
The industry reaction to the scrappage of the Edinburgh supercomputer was strong, and highly critical of the government’s stance, with techUK saying the UK risked "losing out" to other countries in what are crucial industries of the future, and urged the government to make "new proposals quickly".
Regardless of the motivations behind this myriad of summer announcements, the optics for the government and its technology/industrial strategy are not great.
How did we get here?
Funding for the Edinburgh supercomputer was first announced in the March 2023 Budget by then chancellor Jeremy Hunt, after recommendations in the Future of Compute Review for Britian to build an exascale supercomputer. The supercomputer was set to be 50 times faster than any current computers in the UK and would help “researchers model all aspects of the world, test scientific theories and improve products and services in areas such as artificial intelligence, drug discovery, climate change, astrophysics and advanced engineering,”
At the time of its announcement, Sue Daley, director of technology and innovation at techUK, said “Investment in large scale computers is vital for the scientific breakthroughs that will grow our economy and improve our lives.”
Gerard Grech, former CEO of Tech Nation, who had previously welcomed the £1.3bn AI investment fund in 2023 pointed out (via X), in response to its scrapping that “AI and super-computing are two major growth areas that would stimulate investment and economic growth for the UK.”
Why so drastic?
According to a government source quoted in the FT, “the Edinburgh exascale project made little strategic sense given that it was not focused on AI, but on more traditional computing projects such as scientific simulations.”
Ensuring the UK’s sovereign supercomputers are focused on AI tasks makes sense, however, given that the Edinburgh supercomputer hasn’t even been built yet, surely it would have made more sense to simply tweak the specifications accordingly, as opposed to scrapping it and writing off the time, effort and money that has already been spent on it?
Likewise, if Matt Clifford has been tasked with developing the country’s new AI strategy (due to be published this month), why make such drastic decisions before he’s had a chance to put the strategy together? Does the government view technology and artificial intelligence as drivers of economic growth, or not? The Edinburgh supercomputer was a clear indication of the UK’s technology ambitions. Was this simply a political decision to put some clear blue water between the AI strategies of the old and new governments, or was Labour unconvinced of gaining any significant return on the supercomputer’s £800m price tag? With ambitions to have the fastest growth of the G7, why scrap something so lucrative?
What is AI, if not an engine for growth?
Despite the scrappage of such a high-profile project, the government claims to be just as committed to AI. When announcing the appointment of Matt Clifford deliver the AI Opportunities Action Plan, Peter Kyle, secretary of state for science said, “We’re putting AI at the heart of the government’s agenda to boost growth and improve our public services,”
These words are reassuring, so we will have to wait until the Autumn budget in October to see if these words translate into real investment. If the government is serious about delivering the highest growth rate of the G7, then short-term investment programmes in high-growth industries like AI are essential.
In the meantime, all eyes are on the publication of Matt Clifford’s AI Action Plan, which is due out later this month. No pressure Matt…
