Guardian: Fossil Fuel Generators Powering the “Unbelievable” AI Boom are Still “Stranded Assets”

Essay by Eric Worrall

Fumbling the narrative: “if there’s a huge source of energy demand that wants to grow, it will land on these stranded fossil fuel assets.”

‘Just an unbelievable amount of pollution’: how big a threat is AI to the climate?

Ajit Niranjan
Sat 3 Jan 2026 19.00 AEDT

Defenders say AI can do good to fight the climate crisis. But spiralling energy and water costs leave experts worried.

…

“It was jaw-dropping,” said Wilson, a former oil and gas worker from Texas who has documented methane releases for more than a decade and estimates xAI’s Colossus datacentre was spewing more of the planet-heating gas than a large power plant. “Just an unbelievable amount of pollution.”

…

Some experts fear datacentres may derail the shift to a clean economy, adding an unnecessary hurdle to the quixotic task of keeping the planet from heating 1.5C (2.7F). Others are sanguine about the energy costs, arguing they pale in comparison not just to polluting industries, but also to the technology’s power to reshape society.

…

Some of the demand is being met through long-term agreements to buy renewable power …

But for the near future, fossil fuels are set to dominate supply. …

…

“This idea that the lower cost of renewables alone will drive decarbonisation – it’s not enough,” said Daly. “Because if there’s a huge source of energy demand that wants to grow, it will land on these stranded fossil fuel assets.”

…

“Maybe I’m a little naive, but I still believe that AI can do good in terms of fighting the climate crisis – designing the next generation of batteries, tracking deforestation, predicting hurricanes,” …

…

Read more: https://www.theguardian.com/technology/2026/jan/03/just-an-unbelievable-amount-of-pollution-how-big-a-threat-is-ai-to-the-climate

Hannah Daly, who gave us this glorious money quote about stranded fossil fuel assets, in her own words is “a professor in sustainable energy” at University College Cork in Ireland.

Hannah, calling fossil fuel assets which are being used to answer an extraordinary surge in demand “stranded”, it just doesn’t work does it?

So I did the obvious thing and asked an AI to help fix the sentence. Naturally I asked Grok, Elon Musk’s Twitter AI, which seems to be primary villain in the Guardian article.

My question:

This statement seems a little inconsistent: “This idea that the lower cost of renewables alone will drive decarbonisation – it’s not enough. Because if there’s a huge source of energy demand that wants to grow, it will land on these stranded fossil fuel assets.” How can it be reworded to be more consistent?

Grok provided some helpful suggestions.

The apparent inconsistency in the original statement comes from the slightly confusing phrasing around “stranded fossil fuel assets”.

Normally, people use “stranded assets” to mean fossil fuel infrastructure / reserves that become worthless because renewables (or climate policy) have displaced them and demand has fallen.

But the quote is using the term in almost the opposite sense: it suggests that if demand keeps growing strongly (especially from things like AI data centers, electrification, industry, etc.), then even as renewables get very cheap, new demand will still be met — at least partly — by keeping existing fossil fuel plants online longer or building new ones, which prevents (or at least delays) those assets from becoming “stranded” (uneconomic / retired early).

In other words: cheap renewables alone won’t force fossil assets into stranding if overall energy hunger grows fast enough to keep fossil plants useful/profitably running.

Here are several clearer, more internally consistent ways to reword the same idea:

1. Most direct / faithful rephrasing “The idea that the falling cost of renewables alone will be enough to drive full decarbonisation is overly optimistic. If there’s massive new energy demand that wants to keep growing, it will simply continue to be met — at least in part — by existing or new fossil fuel capacity, delaying the stranding of those fossil assets.”
2. Slightly punchier version “Cheap renewables are great, but they won’t automatically decarbonise the system on their own. When there’s enormous, rapidly growing energy demand (think AI, electrification, developing economies), that extra demand will tend to land on whatever capacity is already there — which often means keeping fossil fuel plants running longer instead of stranding them.”
3. Emphasising the demand-side point “Falling renewable costs are powerful, but insufficient by themselves to force deep decarbonisation. If overall energy demand surges dramatically, the market will keep utilising existing fossil infrastructure rather than letting it become stranded — even as new supply increasingly comes from renewables.”
4. Most concise “Cheap renewables won’t decarbonise the grid by themselves. Huge growth in energy demand will simply prolong the life of fossil fuel assets instead of stranding them.”

The core insight the original speaker is trying to convey is usually this: price alone doesn’t win in energy — available volume + reliability + how fast demand is growing also matter enormously. Renewables are winning on new-build marginal cost, but legacy fossil plants can stay in the mix much longer when total demand keeps rising fast. That’s why many energy system experts stress that managing/controlling demand growth (efficiency, flexibility, etc.) is just as critical as cheap clean supply.

Conversation link: https://grok.com/share/c2hhcmQtMi1jb3B5_c3aa4322-880a-4e91-8aef-1533471f7c29

If energy demand surges the way everyone seems to expect, even if renewables were a viable replacement for fossil fuel, how can it possibly still make sense to describe fossil fuel assets as “stranded”? For the foreseeable future, all the “stranded” fossil fuel assets will be running at full capacity to supply the AI boom.

Here’s a thought Hannah – next time you want to criticise AI, why don’t you first ask AI to check what you are saying, to help you fix the obvious mistakes?