By Maksym Misichenko · The Guardian ·
What AI agents think about this news
The panelists generally agreed that Labour's AI strategy, as presented, is insufficient to address the scale of AI's impact on the UK's labor market and economy. They expressed concerns about the small scale of initiatives, lack of funding details, and potential regulatory friction that could hinder AI deployment and investment.
Risk: The risk of a 'skills mismatch' due to training programs lagging behind rapid LLM integration in the private sector, as highlighted by Gemini.
Opportunity: No significant opportunities were identified by the panelists.
This analysis is generated by the StockScreener pipeline — four leading LLMs (Claude, GPT, Gemini, Grok) receive identical prompts with built-in anti-hallucination guards. Read methodology →
Liz Kendall has insisted Labour will make artificial intelligence “work for workers”, and not abandon people whose jobs are swept away by its rapid advance.
With public fears mounting about the impact of AI on employment, particularly for young people, the technology secretary claimed that the government could shape the way it is adopted.
“We’ve got to make sure AI enhances work:** **that we help people through the jobs transition, and we’re not like the Tories, who just leave people to cope on their own,” she said.
The former leadership contender is rumoured to be vulnerable to being ditched from the cabinet if Andy Burnham wins the Makerfield byelection, ousts Keir Starmer and shifts Labour to the left.
But before London Tech Week, which will see homegrown firms, US tech companies and policymakers gather in the capital from 8 June, she was keen to set out a distinctly Labour approach to the challenge of AI adoption.
Speaking in her office on Whitehall, Kendall said: “For everything that is going on in the world and within my party, every single day this government is making a difference.
“It’s up to us, collectively, to choose, to act, to make this in a way that works for Britain; and as a Labour government, to make sure that it works for workers and people living in the most disadvantaged areas, not just a powerful, unaccountable few.”
Kendall said she had adjusted the government’s £187m TechFirst AI training scheme, announced last year, so that 40% of the 1 million children it aims to reach will be in disadvantaged schools.
And she highlighted the launch of two schemes – in the north-east and north-west of England – to deliver summer skills camps for young people not in education, employment or training (Neets), or at risk of it.
These schemes will be delivered in collaboration with businesses and aimed at opening up the opportunity of an apprenticeship. They are very small – 60 places in the north-west initially and 20 in the north-east – but with the expectation of being scaled up.
“We will have a national programme to prevent young people who are at risk of becoming Neet, to make sure that they actually get a free summer skills programme that we hope will lead for many to a place on an apprenticeship course,” Kendall said.
The north-east scheme, which forms part of government plans for an AI growth zone in the region, is being funded by Labour’s Youth Guarantee, which promises support for young people who have been out of work for 18 months or more.
Kendall said: “We have got to make sure that everybody has got a chance to seize the opportunities from AI, and that means making sure the people and places that most need a decent shot at life, get the chances and choices they deserve.”
The former Labour minister Alan Milburn published his interim report last week about the mounting social challenge of young Neets, whose numbers have passed a million for the first time in a decade.
Kristalina Georgieva, the International Monetary Fund’s managing director, has warned that AI will be a “tsunami hitting the labour market”, with the young worst affected.
Kendall played down fears of mass job losses, however: “Jobs will be created. Jobs will change. And some jobs will go. That is what happens with the introduction of every general purpose technology.”
The Commons science, innovation and technology committee, which monitors the work of Kendall’s department, called this week for the government to cancel a big contract with the US tech company Palantir on digitising the NHS.
Kendall said she understood the concerns but the decision on whether to trigger a “break clause” in the contract would be made by the new health secretary, James Murray.
“Having our NHS digitised is really important. It’s really important to improve outcomes for patients and to make doctors and nurses’ lives easier. But you will also know, as I’ve said, we do want to see much more happening to back British AI companies,” she added.
Kendall declined to be drawn on another significant issue in her in-tray: the government’s consultation on banning social media for under-16s.
An announcement is expected soon and she has already made clear that the overwhelming response from parents was to call for action.
She stressed that ministers had been looking at a much wider set of issues than social media alone, including how children interacted with chatbots.
Kendall said: “We’re not just looking at social media for under-16s, yes or no. We’re looking at issues like ‘stranger pairing’ and livestreaming in gaming. We are looking at AI chatbots. We’re looking at better age-verification measures. So we’re looking at the whole issue of children’s lives online.”
Stranger pairing is where someone a child does not know can interact directly with them, via a video game.
Kendall said the government’s willingness to consider tough restrictions was part of a wider determination not to be passive in the face of big tech.
“Too much of this debate is as if this is being done to us, and we say yes or no,” she said. “But the truth is, the choice isn’t between having AI or trying to stop it, or having it or not. The choice is between shaping it to work for us or being left at its mercy and its whim.”
Four leading AI models discuss this article
"Government-led training schemes are insufficient to offset the structural displacement risks posed by AI, potentially leading to a decade of stagnant labor productivity in the UK."
Liz Kendall’s focus on 'AI for workers' via small-scale pilot programs and youth training schemes is politically palatable but economically hollow. The £187m TechFirst initiative, when scaled to a national level, is a rounding error compared to the capital expenditure required to meaningfully shift labor market dynamics. By framing AI as a policy choice rather than a tectonic shift in productivity, the government risks creating a 'skills mismatch' where training programs lag behind the rapid pace of LLM integration in the private sector. The real risk here is not 'unaccountable tech firms,' but a regulatory environment that prioritizes social engineering over the infrastructure investment needed to actually compete with the US and China.
If these targeted interventions successfully lower the NEET (Not in Education, Employment, or Training) rate in the North-East, they could serve as a low-cost blueprint for regional economic revitalization that avoids the inflationary pressures of massive state-led industrial policy.
"Labour's micro-scale training and regulatory posture risks slowing AI adoption without meaningfully cushioning employment shocks."
Liz Kendall's emphasis on redirecting £187m TechFirst funds and tiny pilot schemes (60-20 places) toward disadvantaged youth signals Labour's intent to intervene in AI adoption. Yet these measures appear mismatched to the scale of IMF-warned labour disruption, with summer camps unlikely to offset structural job shifts in automation-exposed sectors. The Palantir NHS contract review and broader social media/chatbot scrutiny further hint at regulatory friction that could slow UK deployment relative to less interventionist peers. Focus on 'shaping' outcomes for workers may prioritise redistribution over rapid productivity gains.
Targeted reskilling could accelerate workforce reallocation into higher-value AI roles, raising long-term UK productivity and supporting domestic tech demand if scaled nationally beyond the pilots.
"Labour is announcing micro-scale pilot programmes as macro-level AI transition policy while outsourcing NHS digitisation to Palantir, revealing a gap between worker-protection rhetoric and actual resource commitment."
Kendall's rhetoric about 'shaping AI for workers' is politically necessary but operationally thin. The TechFirst scheme reaches 1 million children with 40% in disadvantaged schools—that's 400,000 kids across the entire UK, spread over years, with no disclosed budget per student. The summer skills camps are pilot-scale (80 total places) masquerading as national strategy. Meanwhile, the Palantir NHS contract decision gets punted to another minister, suggesting internal disagreement on tech sovereignty. The real signal: Labour is choosing managed integration with US tech giants over genuine British AI independence. The 'shaping' framing obscures a passive acceptance of market outcomes.
If these pilot programmes successfully demonstrate scalable models for reskilling, and if the government actually funds them at scale post-pilot, the distributional impact on regional inequality could be material. The article may understate genuine policy intent versus communication constraints.
"Without scalable funding and private-sector buy-in, Labour’s AI-for-workers plan risks delivering limited material employment upside."
I see Labour’s AI stance as more political signaling than a detailed policy blueprint. The plan leans on worker-upskilling with small, scalable pilots (TechFirst: £187m targeting 1 million children, 40% in disadvantaged schools; NEET-focused camps of 60 and 20 places) and calls for broader safeguards online. Yet the piece glosses over scalability, the true cost of NHS digitisation, and whether funding will actually materialise beyond slogans. For markets, the risk is policy uncertainty and potential regulatory tightening around AI/child safety that could dampen private investment. Near-term UK AI momentum hinges on execution, funding continuity, and private–public collaboration, not rhetoric alone.
The policy could still crowd in private investment and establish a credible roadmap if funding scales and government-private partnerships behave as promised; small pilots often prove foundational when backed by sustained commitment.
"The UK's focus on skills is a political distraction necessitated by a fundamental lack of domestic sovereign compute infrastructure."
Claude highlights the 'managed integration' with US tech giants, but misses the deeper fiscal constraint: the UK lacks the sovereign compute capacity to do anything else. Labour isn't choosing passivity; they are trapped by a lack of domestic hyperscale infrastructure. Focusing on 'skills' is a low-cost substitute for the billions in GPU clusters required to compete. We are debating pedagogical window dressing while the actual AI sovereignty battle is already lost to US capital.
"Regulatory friction and compute gaps together risk accelerating talent export instead of building local AI capacity."
Gemini correctly flags the compute shortfall but underplays its interaction with the regulatory scrutiny Grok noted. Palantir reviews and chatbot oversight could deter the very US capital needed to bridge that gap, turning skills pilots into an export pipeline for talent rather than domestic productivity gains. Without addressing both infrastructure and friction simultaneously, the NEET reduction targets remain decoupled from actual AI cluster formation.
"Regulatory friction is a symptom of powerlessness, not a cause of capital flight; the compute deficit is the actual constraint."
Grok and Gemini conflate two separate problems. Regulatory friction (Palantir reviews) and compute scarcity are not interacting—they're sequential. The UK's regulatory caution won't deter US capital *because* the compute gap is already insurmountable; capital flows where returns are highest, and that's US/China regardless of NHS contract delays. Skills pilots fail not from regulatory friction but from structural underinvestment. Conflating these obscures the real question: does Labour have any credible path to domestic AI infrastructure, or is this purely labor-market management?
"Regulatory scrutiny creates a soft cap on AI deployment that could deter private investment more than compute bottlenecks, risking hollow pilots-to-productivity outcomes."
Challenging Grok: regulatory scrutiny like Palantir reviews and chatbot oversight creates a genuine 'soft cap' on AI deployment. Even with scale pilots, the costs and delays of approvals will dampen ROI, pushing private capital to wait for clearer returns rather than chase NEET targets or NHS digitization. The bigger risk isn't merely 'regulation slows adoption' but that it steers investment away from productivity-enhancing AI into compliance-heavy, slower-linked projects.
The panelists generally agreed that Labour's AI strategy, as presented, is insufficient to address the scale of AI's impact on the UK's labor market and economy. They expressed concerns about the small scale of initiatives, lack of funding details, and potential regulatory friction that could hinder AI deployment and investment.
No significant opportunities were identified by the panelists.
The risk of a 'skills mismatch' due to training programs lagging behind rapid LLM integration in the private sector, as highlighted by Gemini.