Three ways AI has been moulding the global energy market

In this week’s Power Playbook: AI has been moulding the global energy market, influencing how data centre operators approach their budget, wedging its way into Trump’s executive orders and providing a strategic lamppost for the UK’s power grid connections concerns and market development. Let’s break it down.

For energy sector professionals, the conundrum behind AI and its power demand is no secret. Its boon for the energy sector is matched in scale only by the bane of its surging demand placed on the power grid. Which is what makes it so interesting to follow.

And the last two months have demonstrated a rapidly evolving energy market shaped by AI’s influence.

1. Co-located low carbon tech

In late February, UK-based tech company IDTechEx released research exploring why data centres have been investing in nuclear and other carbon-free energy technologies. The report finds that by 2035, the global data centre sector will have saved $150 billion by using low-carbon energy sources compared to a fossil fuel scenario.

The long and short: unprecedented growth in AI creates surging power demand from data centres, leading to hyperscalers wanting new grid capacity with generation methods that align with their espoused sustainability goals. Certain renewables tech offers a cheaper option than that of fossil.

To get more insights on the back of the report’s release, I spoke to Michael Phelan, chief executive and co-founder of GridBeyond:

“It’s driven by the fact that solar and battery probably has the cheapest levelised cost of producing energy today, and wind is not far behind. The Chinese and others have been at this for a good while now, so the costs are already quite low and they will go lower as time goes on.”

More from the Power Playbook:
Trump’s u-turning tariffs: What’s next for a tense energy sector?
Energy innovation and its startups are riding waves – let’s break it down

Indeed, according to IDTechEx, most new solar/wind installations nowadays have an LCOE (levelised cost of electricity) below fossil fuels with tech giants like Amazon, Microsoft, Meta, and Google cited by the report as the leading corporate wind and solar power buyers globally.

IDTechEx also says that emerging energy solutions such as small modular nuclear reactors, hydrogen and fuel cells, enhanced geothermal systems, and grid-scale Li-Ion batteries are now seeing increased investment from the data centre space.

Such options are of course nascent, as said by Phelan: “The other parts – nuclear, hydrogen and fuels cells – need some research and development. They’re not quite there yet…and it will take some time, maybe 10 years, and if people stop investing in technology, it just slows down the time.”

IDTechEx in their report however claims that, considering data centre hyperscalers with money to spend on future-proofing operations sustainably, there are reasons to be confident that investing in such early-stage carbon-free energy solutions now can lead to decreased energy costs in the long term.

2. Trump policy

IDTechEx goes on to say that, with US President Donald Trump’s announcement of a $500 billion investment in AI infrastructure back in January, it is clear that data centre growth will continue. To enable this, new power generation sources will need to come online.

Indeed, AI appears to be on the federal agenda in the US and Trump is not slowing down.

Just last week, Trump issued several executive orders revolving around the impact of AI.

To meet the surging demand placed on the power grid, one concerns grid reliability. Another looks to revive coal as a means of providing a cost-effective and abundant source of energy to meet increasing power demands.

It is thus around AI – at least in rhetoric – that Trump is looking to revive the coal economy, a considerably different angle from IDTechEx’s earlier research tying clean tech growth to meeting AI demands. Two things can, of course, be true at once.

3. UK ‘zombie projects’ cuts

This week in the UK, the country’s energy regulator Ofgem greenlit the National Energy System Operator’s (NESO) plan to reform grid connections to free up space otherwise blocked by ‘zombie’ projects.

The issue has been a pain point for ages, with a lack of access to grid connections holding back projects and new investment in the UK. Some companies have been waiting up to 15 years to be connected.

But what does this have to do with AI? Well, it looks like the country has been shifting its priorities, with AI and data centres one such focus, according to the UK’s Department for Energy Security and Net Zero (DESNZ) in a release.

Under the new changes, connection preference will be given to projects to do with data centres and AI, as well as wind and solar. This also means deprioritising those projects that are not ready or not aligned with the UK’s strategic focuses.

To illustrate, according to the UK’s Department for Energy Security and Net Zero (NZ), new commitments to investing in the UK have topped £38 billion ($50.2 billion) since July 2024 for data centres alone, with grid access the single biggest challenge for these types of projects.

And it would look as though AI may just stand at the helm of where this investment is directed, whether in its application or meeting its demand.

Lowering power demand and Microsoft’s pivot?

I don’t think it’s an exaggeration to claim that AI has been one of the most impactful technologies moulding investment decisions on the global landscape.

However, we stand at a period in time where things are developing at a breathtaking pace.

Take, for example, the impact that the Chinese AI company DeepSeek had at the start of the year. Although many have since called it overhyped, the fact that it required significantly lower energy for its operations was one of many reasons it shook markets. And that we can’t forget.

If AI were to continue to develop in such a way, with lowered power demands a given, what then would happen to the clean tech investment landscape, which as pointed out by IDTechEx focuses heavily on meeting this surging demand.

Then there is Microsoft, which has reportedly delayed development on data centre projects globally as Trump’s tariffs created uncertainty for tech executives.

Said Noelle Walsh, president, Microsoft Cloud Operations, in a LinkedIn post: “Data centre planning is a multi-year and capital-intensive programme we plan for years in advance to ensure we have sufficient infrastructure in the right places.

“In recent years, demand for our cloud and AI services grew more than we could have ever anticipated and to meet this opportunity, we began executing the largest and most ambitious infrastructure scaling project in our history.

“By nature, any significant new endeavour at this size and scale requires agility and refinement as we learn and grow with our customers. What this means is that we are slowing or pausing some early-stage projects.”

Indeed, strategies look to be shifting in a global market defined by impermanence.

Will AI continue to be the centre of gravity around these shifting strategies? I would hazard to guess so.

What about you?

Cheers,
Yusuf Latief
Content Producer
Smart Energy International