A study from the London School of Economics and Systemiq suggests it’s possible to cut global carbon emissions without giving up modern comforts—with AI as our ally in the climate fight.
According to the duo’s research, smart AI applications in just three industries could slash greenhouse gas emissions by 3.2-5.4 billion tonnes each year by 2035.
In contrast to much of what we’ve heard, these reductions would far outweigh the carbon that AI itself produces.
The study, ‘Green and intelligent: the role of AI in the climate transition,’ doesn’t just see AI as a tool for small improvements. Instead, it could help transform our entire economy into something sustainable and inclusive.
Net-zero as an opportunity, not a burden
The researchers suggest we should see the shift to a net-zero economy not as a burden but as “a great opportunity for innovation and sustainable, resilient, and inclusive economic growth.”
They focused on three of the major carbon culprits – power generation, meat and dairy production, and passenger vehicles – which together cause almost half of global emissions. The potential AI savings from just these sectors would more than cancel out the estimated 0.4 to 1.6 billion tonnes of annual emissions from running all those AI data centers.
As the authors put it, “the case for using AI for the climate transition is not only strong but imperative.”
Five big ways AI can help save our planet (and us)
1. Making complex systems smarter
Think about how our modern lives depend on intricate networks for energy, transport, and city living. AI can redesign these systems to work much more efficiently.
Remember those frustrating power outages when the wind stops blowing or clouds cover the sun? AI can help predict these fluctuations in renewable energy and balance them with real-time demand. DeepMind has already shown its AI can boost wind energy’s economic value by 20% by reducing the need for backup power sources.
2. Speeding up discovery and reducing waste
Almost half the emissions cuts needed to reach net-zero by 2050 will rely on technologies that are barely out of the lab today and AI is turbocharging these breakthroughs.
Take Google DeepMind’s GNOME tool, which has already identified over two million new crystal structures that could revolutionise renewable energy and battery storage. Or consider how Amazon’s AI packaging algorithms have saved over three million metric tons of material since 2015.
3. Helping us make better choices
Our daily decisions – from what we eat, to how we travel – could drive up to 70% of emissions reductions by 2050. But making the right choice isn’t always easy.
AI can be our personal environmental coach, breaking down information barriers and offering tailored recommendations. Already using Google Maps’ fuel-efficient routes? That’s AI helping you cut emissions while saving gas money. And those smart home systems like Nest use AI to optimise your heating and cooling, which could save millions of tonnes of CO2 if we all adopted them.
4. Predicting climate changes and policy effects
How do we plan for a changing climate? AI can process enormous datasets to forecast climate patterns with unprecedented accuracy.
Tools like IceNet (developed by the British Antarctic Survey and the Alan Turing Institute) are using AI to predict sea ice levels better than ever before, helping communities and businesses prepare. This capability also extends to helping governments design climate policies that actually work, by learning from countless case studies around the world.
5. Keeping us safe in extreme weather
As climate disasters intensify, early warning can save lives. AI-powered systems for floods and wildfires are becoming essential safety nets.
Google’s Flood Hub uses machine learning to provide flood forecasts up to five days in advance across more than 80 countries. That’s precious time for people to protect their homes and evacuate if necessary.
The numbers support AI cutting global carbon emissions
When researchers crunched the numbers, they found AI could:
- Cut power sector emissions by 1.8 billion tonnes yearly by 2035 just by optimising renewable energy
- Save between 0.9 and 3.0 billion tonnes annually by improving plant-based proteins to taste and feel more like meat
- Reduce vehicle emissions by up to 0.6 billion tonnes each year through shared mobility and better battery technology
Here’s the catch: we can’t just sit back and let market forces determine how AI develops. The researchers call for an “active state” to ensure that AI benefits everyone and the planet.
“Governments have a critical role in ensuring that AI is deployed effectively to accelerate the transition equitably and sustainably,” they conclude.
What this means in practice is creating incentives for green AI research, regulating to minimise environmental impact, and investing in infrastructure so communities worldwide can share in the benefits.
By guiding innovation and working together internationally, we can unlock AI’s full potential to reduce global carbon emissions and tackle the climate crisis—and build a future where both people and the planet can thrive.
