AI, productivity and the future of work: Revisited

Regular readers might remember our Christmas ‘long read’, where we took a thorough look at how generative AI could help solve the UK’s productivity problem - analysing the occasionally unsettling risks and benefits to the current workforce.

Earlier this month we also checked in on what AI means for the future of work more generally, and whether the UK government’s AI Action Plan, released amid much fanfare, was the kind of intervention that could translate into the outcomes we all want: a fitter, happier, more productive economy, powered by fitter, happier, more productive people*.

In between these considered, informative pieces of expert analysis, we also created a mildly snarky haiku about EU AI regulation and drew some parallels between my new puppy and adopting generative AI. You’d be forgiven for thinking that perhaps we don’t take this stuff very seriously, but we really do. And it’s time to check in on how it’s going.

The UK government’s latest move(s)

The UK government's launch of an AI Playbook on 10 February 2025 is another big step forward in guiding everyone - particularly public sector organisations - through the safe and effective implementation of AI technology. It’s comprehensive, it builds upon the previous framework and it’s honestly Very Good.

Which is a shame in some ways. It’s much easier to write an engaging blog about what’s wrong. Criticising governments is often all too easy. And fun.

The following day, the UK caused a bit of a stir by not signing up to an international declaration on AI use (the US didn’t sign up either, but that was less surprising). They cited concerns around global data security and governance.

That decision is harder to weigh up: we like committing to things being safe, ethical and sustainable, but the challenge is always going to be in how not what. How do we police new rules to avoid them being meaningless? How are we monitoring everyone, incentivising and rewarding good behaviour and sanctioning transgressions without stifling innovation and growth – and without expending public resources in the process.

There is no area of public policy where that question doesn’t apply: a government’s job is to pick where to strike a balance. Traditionally, UK governments – of any flavour – tend toward loitering in a grey area, generating rules as precedent emerges, until it becomes essential to draw a line. And that’s why we’re seeing different approaches emerge either side of the English Channel: not everyone has that risk appetite.

Progress!

AI is already beginning to dramatically transform industries and job roles – highlighting the urgent need for workers to adapt and upskill. The AI Opportunities Action Plan emphasised skills development and job creation in the AI sector. The new AI Playbook also supports these goals by offering detailed guidance and principles for AI deployment. And they’re also Very Good. Pop it on your reading list, if you haven’t already.

The playbook outlines ten essential principles for using AI lawfully, ethically, and responsibly, establishing meaningful human control and managing the AI lifecycle. It provides clear explanations of AI technologies, real-world use cases, and corporate guidance for policy professionals and decision-makers. Used sensibly, this really will support the public sector in understanding AI's capabilities and limitations, mitigating risks, and driving innovation.

…contrasting with EU AI regulation

The UK's approach to AI regulation contrasts with the EU's more risk-averse approach. That means that by fostering an environment that encourages innovation (while actively taking steps to uphold ethical standards) the UK clearly aims to position itself as a global leader in AI development. This approach could potentially accelerate AI adoption and innovation, benefiting all sectors and boosting the UK economy.

So… job done?

Challenges, as they say, remain. The skills gap continues to be a pressing issue with no serious investment announced to narrow it, in sharp contrast with the level of investment in physical tech infrastructure. The AI Playbook provides for free e-learning courses and establishes an Artificial Intelligence Community of Practice for knowledge sharing, but the distance between strategy and action is a worry, particularly for SMEs, charities, and public services with limited resources and even more limited capacity to pivot their workforce. Mind the gap.

To bridge the divide between strategy and action, it will be essential to support SMEs and cash-poor organisations in adopting AI technologies. That includes providing accessible training and resources, fostering collaboration between academia, industry, and government, and making sure that AI benefits are distributed equitably across all different regions and groups of people. If anyone in the public sector asks you what’s needed, we suggest that’s one answer. But, if you’re in the UK, take a second to thank the nearest civil servant for not starting with a pile of legislation and controls.

What’s next?

The UK government's AI Playbook is a positive step, looking at practical ways to maximise its potential for public good while addressing the risks. But this is about all the choices we make for the organisations we run: invest in skills development, collaborate, think strategically about what you need, invest responsibly and give people starting their careers good advice, with an eye on how skills needs are changing. And don’t forget to have fun.

*Yes, that’s a Radiohead reference. A treat for the GenX-ers.

‍