Will AI cut more emissions than it causes?
The claim: “AI will cut more emissions than it causes.”
Nobody knows yet. The IEA estimates that if existing AI tools were widely used, for example in factories, buildings and transport, the cuts could equal around 5% of energy-related emissions in 2035: three to four times what data centres would emit. That is a scenario, not a measurement. The IEA says there is “currently no momentum” behind that uptake, that rebound effects could undercut the gains, and that AI is “not a silver bullet”. Data centres’ own emissions are already happening.
The bigger pictureData centres’ own emissions, from the electricity they consume, were about 180 million tonnes of CO₂ in the IEA’s 2025 report, heading for 300 Mt by 2035 in its Base Case and up to 500 Mt in its Lift-Off Case. That stays below 1.5% of energy-sector emissions.
Why the claim misleads
It states a forecast as if it were a result. The savings depend on wide uptake that isn’t happening yet; the emissions from data centres are already real.
What nobody knows yet
- Nobody has measured how much CO₂ AI has actually avoided worldwide so far.
- The IEA doesn’t put a size on rebound effects.
- Other studies give their own estimates; Hum hasn’t yet checked them on the page.
Sources (10)
- IEA, Energy and AI (2025) Forecast
“We estimate that emissions reductions from the broad application of existing AI-led solutions to be equivalent to around 5% of energy-related emissions in 2035.”
Quote checked on the page, 25 Oct 2026 - IEA, Energy and AI (2025) Forecast
“Emissions from electricity use by data centres grows from 180 million tonnes (Mt) today to 300 Mt in the Base Case by 2035, and up to 500 Mt in the Lift-Off Case.”
Quote checked on the page, 25 Oct 2026 - IEA, Energy and AI (2025) Forecast
“While these emissions remain below 1.5% of the total energy sector emissions in this period, data centres are among the fastest growing sources of emissions.”
Quote checked on the page, 25 Oct 2026 - IEA, Energy and AI (2025) Forecast
“The widespread adoption of existing AI applications could lead to emissions reductions that are far larger than emissions from data centres – but also far smaller than what is needed to address climate change.”
Quote checked on the page, 25 Oct 2026 - IEA, Energy and AI (2025) Forecast
“Rebound effects – for example from modal shifts away from public transport to autonomous cars – could undercut some of these benefits.”
Quote checked on the page, 25 Oct 2026 - IEA, Energy and AI (2025) Forecast
“AI can be a tool in reducing emissions, but it is not a silver bullet and does not remove the need for proactive policy.”
Quote checked on the page, 25 Oct 2026 - IEA, Energy and AI (2025) Forecast
“Concerns that AI could accelerate climate change appear overstated, as do expectations that AI alone will address the issue”
Quote checked on the page, 25 Oct 2026 - IEA, Energy and AI (2025) Forecast
“These potential emissions reductions, if realised, would be three times larger than the total data centre emissions in the Lift-off Case, and four times larger than those in the Base Case.”
Quote checked on the page, 25 Oct 2026 - IEA, Energy and AI (2025) Forecast
“It is vital to note that there is currently no momentum that could ensure the widespread adoption of these AI applications.”
Quote checked on the page, 25 Oct 2026 - IEA, Energy and AI (2025) Forecast
“Therefore, their aggregate impact, even in 2035, could be marginal if the necessary enabling conditions are not created.”
Quote checked on the page, 25 Oct 2026
Cite this
IEA, Energy and AI (2025), as quoted in Hum, “Will AI cut more emissions than it causes?”, https://humearth.org/q/can-ai-cut-emissions (quote checked 25 Oct 2026).
Cite the original source first; Hum is where you found it.
What do Wh, kWh and TWh mean?
Wh is a watt-hour: a 1-watt light left on for an hour. 1 kWh = 1,000 Wh, and 1 TWh = 1 billion kWh. Hum compares each number with something everyday so you don’t need the units.
Send this answer
Will AI cut more emissions than it causes? In an IEA scenario, yes: around 5% of energy-related emissions in 2035 if AI tools are widely used. But that’s a forecast, and the IEA calls AI “not a silver bullet”.