August 31, 2026

Britain Wants AI. Who Gets the Power?

Last week, a client asked me a version of the same question I keep hearing in different outfits: if the UK is so serious about AI, why does building the plumbing feel like smuggling a nightclub into a monastery?

Fair question.

A proposed mega data centre in Scotland, reported by TechRadar, has hit what was described as huge opposition. And it is not an isolated punch-up. Similar resistance is cropping up across the UK, from concerns over energy use and water consumption to noise, land use and whether local communities actually get anything useful in return. Source: TechRadar, 'World’s second-biggest data center proposed in Scotland meets huge opposition — and it's part of a wider trend across the United Kingdom'.

Here is the bit nobody should pretend is surprising. You cannot spend months talking about sovereign AI capability, digital transformation and economic competitiveness, then act shocked when people ask where the power comes from, who gets the jobs, and why the giant beige warehouse has landed next to them like an alien fridge.

My view is simple. The UK does not have a tech ambition problem. It has an execution credibility problem.

AI without infrastructure is just PowerPoint

There is a particular kind of executive optimism that says: we need AI, AI needs data centres, so obviously data centres will appear.

That is toddler logic with a budget.

Data centres are not abstract 'digital infrastructure'. They are physical, power-hungry, politically visible assets. They need land, planning approval, grid capacity, cooling, security, fibre, resilience and people who know what they are doing. Preferably all at once. Preferably yesterday.

The UK government wants AI-led growth. Fine. But AI at scale is not built on vibes, keynote slides or someone shouting 'innovation' in Canary Wharf. It is built on racks, substations, network architecture, and infrastructure teams who can keep the lights on while everyone else posts thought leadership on LinkedIn.

If local opposition slows projects, the impact is not just delayed construction. It ripples into cloud strategy, cyber resilience, sovereign hosting, cost models and, yes, hiring. Because companies do not just need facilities. They need the people to design, run and secure them.

The real issue is trust, not technology

Most local resistance is framed as anti-growth or anti-tech. That is lazy.

People are usually not objecting to machine learning as a concept. They are objecting to poor trade-offs, vague promises and the suspicion that they get the inconvenience while somebody else gets the upside.

Again, not irrational.

If you are a local resident and someone says, 'Good news, a hyperscale facility is coming', your next questions are obvious:

• What does this mean for energy use in my area?

• What happens to water resources and local ecology?

• How many meaningful local jobs does this actually create?

• What strain does it put on transport and utilities?

• Why here?

Too often, developers and policymakers answer those questions like they are reading from a hostage note.

This is where the whole thing falls apart. When communication is weak, communities assume the worst. And to be blunt, they often have reason to. For years, 'development' has too often been sold like a magic bean, with benefits talked up and costs quietly parked under the carpet.

You cannot build strategic infrastructure on public scepticism and then complain that the public is sceptical.

There is a hiring story buried inside this fight

This is the bit I care about professionally, because recruitment tends to be where strategy finally meets reality.

Every time a major digital infrastructure project is delayed, accelerated, relocated or politically complicated, three things happen in the talent market.

Demand gets distorted

Businesses rush to secure Infrastructure, Cloud, Network, Cyber and Data talent before plans are fully settled. That creates panicky hiring, odd role specs and compensation benchmarks pulled from fantasyland.

I have seen companies ask for a hybrid of data centre operations lead, cloud architect, security strategist and programme manager, then wonder why the shortlist looks suspiciously imaginary.

That person does exist, by the way. They are already employed and not terribly interested in your 'competitive package'.

Good candidates get more selective

The best people are not just choosing salaries. They are choosing credibility.

If a company’s infrastructure roadmap feels vague, politically exposed or obviously undercooked, strong candidates hesitate. Top technical talent can smell chaos from across the M25.

They ask themselves:

• Is this project real?

• Is leadership aligned?

• Will I have budget and authority?

• Am I joining a serious build, or a future post-mortem?

If you cannot answer those cleanly, do not expect easy hiring.

The skills gap becomes more expensive

When the market senses urgency, prices move. Not just salaries, but time-to-hire, counteroffers, consultant spend and the hidden cost of indecision.

This matters because the UK already has a tight market for experienced infrastructure and cyber professionals. Add a wave of AI-adjacent infrastructure demand, and suddenly everyone is fishing in the same small pond while insisting their bucket is unique.

It is not unique. It is a bucket.

What smart leaders should do now

If you are a founder, CTO, COO or people leader watching this trend unfold, here is the practical bit.

Do not wait for the planning wars to settle before sorting your talent strategy. By then, the best people will be gone and the market will be even noisier.

Treat infrastructure hiring as strategic, not reactive

If AI, cloud modernisation, resilience or sovereign hosting matters to your business, your hiring plan should reflect that now.

Ask:

• Which roles are genuinely mission-critical over the next 12 to 24 months?

• What capabilities can we build in-house versus buy via partners?

• Where are we over-specifying roles and scaring off strong candidates?

• What is our back-up plan if projects slip?

Waiting until the pressure is obvious is exactly how businesses end up making six-figure mistakes with three rounds of interviews and a lot of brave faces.

Sell the mission properly

The best candidates want to know why the work matters.

'Join us, we are doing exciting things in AI' is not a pitch. It is wallpaper.

A better version sounds like this:

• Here is the infrastructure challenge we are solving

• Here is why it matters commercially

• Here is the complexity involved

• Here is the authority you will have

• Here is how leadership is backing it

Serious people are attracted to serious clarity.

Build credibility before you need it

If your business relies on infrastructure-heavy growth, candidates will scrutinise your maturity.

So sharpen the basics:

• Clear reporting lines

• Defined technical ownership

• Realistic project timelines

• Thoughtful salary benchmarking

• A hiring process that does not feel like a medieval obstacle course

Nothing kills momentum faster than a business claiming urgency and then taking five weeks to schedule a second interview.

That is not rigour. That is administrative jazz.

The UK needs a grown-up conversation

There is a wider point here.

The country needs to decide whether it wants the benefits of AI and digital sovereignty enough to deal honestly with the infrastructure consequences.

That does not mean waving through every data centre proposal with a patriotic grin. Some developments will be badly planned, badly located or badly justified. Opposition is not automatically obstruction.

But if every region says no, every planning process drags, and every project gets treated like a scandal in waiting, then the UK’s AI ambitions become performative. We will still buy the software, still consume the compute, still talk a great game. We will just rely more heavily on infrastructure elsewhere and wonder why strategic autonomy feels a bit theoretical.

And from a labour market perspective, that uncertainty is poison. It makes long-term workforce planning harder. It weakens employer confidence. It pushes strong talent towards firms and regions that feel clearer, faster and more serious.

A simple framework for leaders

If you are trying to make sense of this trend internally, use this quick filter:

The GRID test

• Growth: What growth bets depend on infrastructure capacity?

• Risk: What happens if deployment, hosting or resilience plans are delayed?

• In-house: Which capabilities must we own ourselves?

• Demand: What roles will become hardest to hire if the market tightens further?

If you cannot answer those four cleanly, your talent plan is probably too reactive.

Conclusion

Here is the blunt truth. Britain cannot muscle its way into AI leadership while treating the infrastructure underneath it like an awkward detail. Data centres are not sexy, but neither are foundations, and you still need them before building the penthouse.

The businesses that win will be the ones that grasp this early. They will plan for infrastructure friction, hire ahead of the bottlenecks, and talk about digital ambition like adults, not magicians.

If your roadmap depends on cloud, data, resilience or AI, now is the time to get your hiring story sorted. Because when the market tightens and the planning rows get louder, hope is not a strategy. It is just expensive optimism in a nicer blazer.

Back to news