Last week, all eyes were on the Chancellor as she confirmed the government’s support for a third runway at Heathrow during a wide-ranging speech focused on unlocking growth in the UK economy. Announcing the decision, Rachel Reeves declared, “Low growth is not our destiny, but growth will not come without a fight — without a government willing to take the right decisions now to change our country’s future for the better.”
These words sent a clear signal to anyone who wasn’t already aware: growth will be the government’s fundamental mission — its raison d’être — seemingly at any cost.
Unsurprisingly, this has sparked concerns among prominent figures in the net zero space, who worry that the UK’s broader climate ambitions could be undermined by an all-encompassing push for growth.
However, amid concerns about the tension between growth and net zero, one crucial factor remains largely overlooked but is fundamentally tied to both: Artificial Intelligence.
Could AI Undermine the Government’s Net Zero Strategy?
In January, the Prime Minister unveiled the government’s AI Opportunities Action Plan, an ambitious strategy aimed at using artificial intelligence to ‘unleash growth’ across the UK economy. Central to this plan is a twentyfold increase in AI computing power by 2030, with the development of ‘AI growth zones’ to accelerate planning approvals and facilitate the construction of more AI data centres.
However, while AI holds immense potential for economic growth and technological advancement, it comes with a significant challenge—its enormous energy consumption. A single ChatGPT query, for example, uses ten times more energy than a Google search. This is because AI data centres require massive amounts of electricity to run, cool, and maintain their systems.
To put the scale of this challenge into perspective, AI and related digital infrastructure consumed around 3.6 terawatt-hours (TWh) of electricity in the UK in 2020. If the government’s target of a twentyfold increase in AI computing power is realised, this could surge to 72 TWh by 2030—accounting for over a quarter of the UK’s total electricity consumption in 2021 (approximately 261 TWh).
A recent study by Loughborough University highlights the impact of this projected demand increase, suggesting that energy consumption from the digital data sector could outpace the growth of renewable energy capacity in the UK. Even John Pettigrew, head of National Grid, has warned that demand from commercial data centres alone is expected to increase sixfold by 2035.
This looming energy challenge raises a critical question: Has the government fully considered AI’s impact on the UK’s net zero targets?
The Department for Energy Security and Net Zero has made some positive noises in this area, suggesting that the increased adoption of AI in the energy sector could not only drive groundbreaking innovations to combat the climate crisis but also enhance the efficiency of the UK’s energy network.
However, while AI can be a powerful tool for assisting the transition — enhancing energy efficiency, optimising grid management, and advancing green innovations — its unchecked expansion could risk undermining the very sustainability goals the UK has pledged to meet.
What is the solution?
There are reasons to be optimistic. The emergence of DeepSeek, a Chinese AI model similar to ChatGPT but far more efficient, suggests that advances in AI technology could lead to models that deliver equal or superior performance while consuming significantly less power.
However, relying solely on future technological improvements is a risky strategy.
If the UK is serious about meeting its net zero commitments while embracing AI-driven growth, the government’s net zero strategy must explicitly address AI infrastructure. This means setting clear policies to ensure AI expansion is powered by clean energy, incentivising the development of low-carbon data centres, and placing greater emphasis on energy efficiency in AI research.
Collaboration between the government, AI industry leaders, and the energy sector will be essential. The government’s proposed AI Energy Council, chaired by the Science and Energy Secretaries, is a welcome first step in this area. This council will work with energy companies to understand the energy demands and challenges that will accompany AI development.
However, there is a risk of kicking the can down the road. We already have a clear understanding of the energy demands of existing AI infrastructure. Failing to act now could result in an AI revolution that comes at the expense of our net zero ambitions.Find out more about our work on net zero
