What AI and the Climate have in common
- Matteo Deidda
- Jan 12
- 4 min read
I just finished a three-week course on the Ethics of AI from the London School of Economics. As someone who's spent 15 years working on climate change, I kept noticing parallels.
Not because AI and climate are the same challenge. They're not. But because the dynamics around both feel remarkably similar, and I think there are some useful lessons we can draw from one to inform how we approach the other.
The technology is neutral. What we do with it isn't.
AI, like fossil fuels before it, is fundamentally a tool. The question isn't whether the technology is inherently good or bad. The question is: how do we use it, who controls it, and what guardrails do we put in place?
Right now, AI development is concentrated in the hands of a few massive companies. That dynamic feels familiar if you've worked in climate. It's similar to what we've seen with oil and gas: a small group of actors with enormous influence, shaping markets, moving faster than regulation, and building systems that become harder to change the more entrenched they become.
With fossil fuels, we spent decades letting the industry self-regulate. Now we're trying to retrofit policy onto deeply embedded infrastructure and business models. It's difficult work, and progress is slower than we'd like.
We have an opportunity not to repeat that pattern with AI. But it requires making choices now, while we still have flexibility.
AI brings both opportunity and risk
The potential benefits of AI are real and significant: breakthroughs in healthcare, better climate modelling, efficiency gains across industries, new ways to solve complex problems.
But the risks are equally real. AI can disrupt labor markets at scale. It can deepen inequality if the benefits flow primarily to those who own the technology. It can affect democratic processes through misinformation, surveillance, and algorithmic bias. And it can do all of this much faster than climate change has unfolded.
The conversations about Artificial General Intelligence (AGI), systems that can match or exceed human intelligence across all tasks, have shifted from "if" to "when." Most experts now think we're talking years, not decades.
That's a short timeframe to figure out governance, accountability, and societal adaptation. Not a reason to panic, but definitely a reason to engage seriously with what's coming.
What climate experience teaches us about managing transformative technology
Working in climate for 15 years has given me some perspective on how societies handle big, complex changes. Here's what feels relevant to AI:
It's easier to build in safeguards early than retrofit them later.
With climate, we're now trying to decarbonise economies while they're running. It's complex, expensive, and politically challenging. A lot of infrastructure and business models were built without climate considerations baked in, and now we're working around those constraints.
AI is still in its early stages. This is the time when ethical frameworks, accountability mechanisms, and governance structures are easier to integrate, before the technology becomes so embedded that changing course becomes much harder.
Global challenges need coordinated responses.
Climate doesn't respect borders, and neither does AI. Regional regulation is useful, but it's incomplete if other parts of the world take different approaches. The challenge is that AI development is competitive, between companies and between countries. That tension between cooperation and competition is familiar from climate negotiations.
Self-regulation has limits.
Companies building transformative technologies don't always see the full range of risks, and they're optimising for different things, speed, capability, market position. That's not necessarily malicious, it's just structural. We learned from climate that voluntary action is valuable but insufficient. At some point, you need clear rules and accountability.
Why this matters for sustainability professionals
If you work in climate or sustainability, you might be wondering why AI should be on your radar.
A few reasons:
AI is already being used in sustainability work. Carbon accounting, supply chain transparency, climate modelling, energy optimisation, AI is increasingly part of how we do this work. Understanding its capabilities and limitations makes you more effective.
AI governance is heading toward ESG functions. Just like sustainability moved from a niche concern to a board-level issue, AI ethics and impact assessment are moving in the same direction. The skills overlap is significant: systems thinking, risk assessment, stakeholder engagement, navigating uncertainty.
The field is evolving quickly. If you're interested in staying current in sustainability, understanding how AI intersects with your work is becoming less optional. Not because you need to become a data scientist, but because the tools and context are changing.
We're writing the story now
The outcome of AI development isn't predetermined. We're still early enough that choices matter, about regulation, about governance, about how we deploy the technology and who benefits from it.
Working in sustainability taught me that these windows don't stay open forever. Not because it's too late to act, but because the later you intervene, the harder it gets and the more constrained your options become.
So what does that mean practically?
Support thoughtful regulation. Not just principles, but enforceable frameworks with real accountability.
Push for international coordination. AI governance works better when it's aligned across regions, even if perfect coordination isn't realistic.
Keep learning. If you work in sustainability, spending some time understanding AI, what it is, where it's going, how it connects to your work—is probably worth it.
The parallels between AI and climate aren't perfect, but they're instructive. Both involve powerful, transformative technologies. Both require us to think carefully about governance and equity. And both benefit from learning from past experience rather than repeating the same patterns.
I don't think we're doomed to make the same mistakes with AI that we made with climate. But I do think we need to be intentional about learning those lessons.
*If you're interested in the Ethics of AI, the LSE course was excellent. It's given me a lot to think about.hips. These are just things I've bought with my own money and actually use.
Comments