“We can’t fight climate change without AI.” This powerful narrative is currently shaping the conversation around AI and climate change, offering a seductive promise of technological salvation. At the same time, environmental activists and researchers have been raising a critical point: The very tools we’re counting on to solve the climate crisis may also be accelerating it. The relationship between AI and climate seems to be more complex than our public debate currently reflects. So I decided to dig deeper.

I try to live sustainably – I bring my own cup, I bike, I buy local. And yet, I love technology. I use AI tools, I stream videos, I work in the cloud. So lately, I started to question my own efforts: Is my digital life actually outpacing my green one?

I’m deeply grateful that environmental concerns have reached the center of public debate. Even despite recent backlashes, sustainability remains on the agenda — in politics, business, education, and culture. It’s easy to forget how recent that shift really is. We owe that progress to the tireless work of movements and activists who pushed these issues into the spotlight. But when we examine what’s actually being done, implementation rarely matches the rhetoric. While sustainability is frequently named, it’s seldom prioritized. It’s acknowledged in speeches and mission statements, yet routinely sidelined in practice. 

This raises a troubling question: Are we building AI’s future on the same flawed sustainability foundations? To be fair, there are growing efforts to make AI more sustainable — and they deserve recognition. But in practice, sustainability often feels like a footnote rather than a driving force behind innovation – especially when it comes to the major players on the field.

Recent developments in AI bring the question of energy efficiency into sharper focus. Take the new ChatGPT voice Monday — sassy, fun, and designed to entertain. It quickly led to a wave of users chatting just for the experience.¹ AI-enhanced search engines like Google raise similar questions. What used to be a quick, low-energy query now runs through large-scale models — making even simple searches significantly more resource-intensive.²

I couldn’t help but wonder: What kind of digital future are we building?

AI and Climate Change – The Climate Hero We Hoped For?

The statement at the beginning of this article is, in many ways, true. Digitalization isn’t just part of the climate conversation — it’s shaping it. AI has shown impressive potential — especially when it comes to making systems smarter, leaner, and more energy-efficient. Across industries, digital innovation is being used to reduce emissions, cut waste, and accelerate climate solutions. Here are just four examples: 

• AI technologies like DeepMind‘s optimization of wind energy have demonstrated a 20% boost in renewable energy economic value.
• Google Maps‘ AI-driven eco-friendly routing has helped reduce 1 million tons of CO2 annually by optimizing traffic routes in cities.³
• AI enables more sophisticated climate modelling and simulations, helping policymakers understand complex climate systems and simulate potential impacts of policies.⁴
• Even remote work, made possible by digital infrastructure, led to a CO2 reduction of around 54 million tons globally in 2020.⁵

These are just a few examples – other promising use cases include AI for forest protection, waste sorting, ocean plastic mapping, and decarbonizing heavy industry. Used wisely, digital tools can support our transition toward sustainability. In fact, they are already contributing to climate solutions — and their potential is far from exhausted. It’s important to note though that many of the statistics and claims surrounding AI and climate change focus on the potential of the technology rather than its current impact. While the future of AI in climate action is promising, much of the data still reflects what AI could achieve, rather than what it has already done at scale.

The Not-So-Green Reality 

But, the very tech we rely on to “save” us is also part of the problem. AI and its development relies on massive infrastructure: data centers, servers, cooling systems – all consuming huge amounts of electricity and water. Companies like Meta and Google are even considering building nuclear power plants to meet AI’s growing demands.This demand for energy, cooling, and space often comes at the expense of local ecosystems. And despite their massive electricity consumption, AI applications like autonomous vehicles and delivery drones risk disrupting wildlife and fragile habitats. Lastly, let’s not forget the growing e-waste problem — one we’ve yet to find truly sustainable solutions for. The numbers speak for themselves:

• Training GPT-3 emitted around 500 tons of CO2.⁶ While that may not sound like much — especially compared to global emissions — it’s just one model, trained once. Multiply that by thousands of models, millions of users, and the energy-intensive infrastructure they rely on, and it becomes harder to ignore..
• Data centers alone already consume about 1-1.5% of global electricity.⁷
• The ICT sector’s emissions now rival those of the entire airline industry – and it’s growing quite rapidly.⁸
• E-waste exceeded 59 million tons in 2022, most of it unrecycled.⁹

So while we build smarter tech, we’re also creating massive, invisible energy and water demands — hidden from view, but carrying real consequences for ecosystems, wildlife, and local communities. Many of these questions — like what to do with mounting e-waste — remain unanswered. Meanwhile, development is moving faster than our ability to fully grasp its consequences.

The Power Behind the Tools

Technology is never developed in a vacuum. It’s shaped by economic interests, political power, and the values of those who design, own, and control it.

The environmental question around AI isn’t just about energy efficiency or carbon footprints — it’s about who decides how these systems are built, what gets optimized, and what gets ignored. Is sustainability a core metric in development, or just a checkbox in a CSR report? Are the benefits and burdens of AI shared equitably — or do they mirror existing power imbalances?

“We’re building a dystopia just to make people click on ads.”¹⁰ With this quote, Zeynep Tufekci sharply — and more accurately than most — captures the economic logic that drives and shapes much of today’s digital trends. Platforms like Google and Facebook are engineered to keep users scrolling and clicking — optimizing for attention, not for sustainability. The environmental consequences of that design — from energy-hungry data centers to always-on algorithms — are routinely sidelined.¹¹ Ultimately, the answer to whether technology helps or harms the planet may have less to do with the tools themselves, and more to do with who controls them — and why.

Technological Evolution in an Age of Global Conflict

One of the biggest challenges facing technological progress today is the volatile geopolitical landscape in which it unfolds. We’re witnessing unprecedented digital innovation during one of the most unstable periods in recent history: the rise of authoritarianism, escalating trade wars, and the highest number of violent conflicts since World War II. In conflicts like the war in Ukraine and Israel’s military offensive in Gaza AI is already being deployed on the battlefield — largely outside public scrutiny, but with devastating consequences, as tragically illustrated by the scale of destruction in Gaza.¹²

At the same time, an AI arms race is accelerating between major global powers, particularly the United States and China — two countries already entrenched in economic rivalry.¹³ The decline of multilateral cooperation and the weakening of international governance mechanisms only deepen the challenge: we lack meaningful and binding global standards for responsible technological development.

Meanwhile, as previously outlined, the tech sector is operating with increasing freedom. Regulatory safeguards are being rolled back, giving private industry more latitude than ever — and placing critical decisions about the future of AI into the hands of a few powerful actors.¹⁴ And although companies brand themselves with terms like “open AI,” in reality much of this technology is being shaped behind closed doors — not through public debate, but by a small group of powerful tech executives whose influence extends into politics, global policy, and even military strategy. In this environment, climate concerns aren’t just deprioritized — they’re systematically sidelined.

And while it’s absolutely true that we already possess many of the tools, data, and solutions to address environmental challenges, my conclusion is sobering: At this point, there simply isn’t enough political will or power to ensure a climate-positive — or even climate-neutral — development of AI and emerging technologies. I believe in technology’s potential to transform our world — deeply and earnestly. But that transformation must be guided with purpose, with sustainability at its core rather than as an afterthought. And right now, I fear we’re drifting off course.

So… What Kind of Future Are We Building?

With no Planet B, this isn’t someone else’s problem to solve. It’s ours — yours and mine. The engineers. The policymakers. The activists. The consumers. The citizens. To create a sustainable future, we need to ask tough questions about AI and climate change — and ensure that innovation doesn’t come at the planet’s expense. What we need now are radical governance innovations, renewed international cooperation, and a democratic movement of people reclaiming their voice in how technology is developed — and for whom.

So I’m asking you directly: What do you think? Is our current technological path helping us heal the planet — or accelerating its decline? Where do you see hope? What keeps you up at night? What solutions are we overlooking? Challenge my perspective. Show me what I’m missing. Share your expertise, your doubts, your vision. Because, as the saying goes, “the best way to predict the future is to create it.” And we are. We are coding, designing, and deploying it right now. The question is: will we have a say in where it leads us?

It’s time to rethink the future we’re building.

References

Tong, Anna, and Akash Sriram. “OpenAI Unveils New AI Model as Competition Heats Up.” Reuters, 14 May 2024, https://www.reuters.com/technology/openai-announce-chatgpt-product-improvements-monday-2024-05-13/ ^↩
Parshall, Allison. “What Do Google’s AI Answers Cost the Environment?” Scientific American, 11 June 2024, https://www.scientificamerican.com/article/what-do-googles-ai-answers-cost-the-environment/.^↩
Stern, Nicholas, and Mattia Romani. “AI’s Role in the Climate Transition and How It Can Drive Growth.” World Economic Forum, 16 Jan. 2025, https://www.weforum.org/stories/2025/01/artificial-intelligence-climate-transition-drive-growth/.^↩
United Nations University | Institute for Environment and Human Security. “5 Insights into AI as a Double-Edged Sword in Climate Action.” United Nations University, 7 Aug. 2024, https://unu.edu/ehs/series/5-insights-ai-double-edged-sword-climate-action.^↩
Anderson, Kara. “Remote vs Office: Which One Is Greener?” Greenly.earth, 15 Jan. 2024, https://greenly.earth/en-gb/blog/company-guide/remote-vs-office–which-one-is-greener.^↩
Plan Be Eco. “AI’s Carbon Footprint – How Does the Popularity of Artificial Intelligence Affect the Climate?” Plan Be Eco, 8 May 2024, https://planbe.eco/en/blog/ais-carbon-footprint-how-does-the-popularity-of-artificial-intelligence-affect-the-climate/.^↩
Dhanabalan Thangam, et al. “Impact of Data Centers on Power Consumption, Climate Change, and Sustainability.” Advances in Computational Intelligence and Robotics Book Series, 15 Mar. 2024, pp. 60–83, https://doi.org/10.4018/979-8-3693-1552-1.ch004.^↩
International Telecommunication Union (ITU). “Measuring the Emissions & Energy Footprint of the ICT Sector: Implications for Climate Action – ITU.” ITU, 13 Aug. 2024, https://www.itu.int/hub/publication/d-ind-clim-2023-01/.^↩
World Health Organization (WHO). “Electronic Waste (E-Waste).” World Health Organization, World Health Organization, 1 Oct. 2024, https://www.who.int/news-room/fact-sheets/detail/electronic-waste-(e-waste).^↩
Zeynep Tufekci. “We’re Building a Dystopia Just to Make People Click on Ads.” Ted.com, TED Talks, 2017, https://www.ted.com/talks/zeynep_tufekci_we_re_building_a_dystopia_just_to_make_people_click_on_ads?language=en.^↩
Beyond Fossil Fuels. “New Data Centres Could Undermine Europe’s Energy Transition, Eating into Its Emissions Cuts, Study Finds.” Beyond Fossil Fuels, 9 Feb. 2025, https://beyondfossilfuels.org/2025/02/10/new-data-centres-could-undermine-europes-energy-transition-eating-into-its-emissions-cuts/.^↩
Serhan, Yasmeen. “How Israel Uses AI in Gaza—and What It Might Mean for the Future of Warfare.” TIME, Time, 18 Dec. 2024, https://time.com/7202584/gaza-ukraine-ai-warfare/.^↩
Lee, Kai-Fu. AI SUPERPOWERS : China, Silicon Valley, and the New World Order. S.L., Mariner Books, 2018.^↩
Elliott, Vittoria. “What Project 2025 Means for Big Tech … and Everyone Else.” WIRED, Aug. 2024, http://www.wired.com/story/project-2025-tech-industry/. Accessed 6 Apr. 2025.^↩