From Scroll to System Change: How Youth Are Rewriting Climate Action with AI
We are the generation that grew up with technology in our hands. We scroll through climate headlines before breakfast, experiment with artificial intelligence (AI) tools between classes and work, and build digital communities that cross borders in seconds. Technology is not something we “adopt” later in life, it is something we live with.
Yet when it comes to moments of crisis, new technologies are often pushed into a simple box: hero or villain. But it is not that easy. A World Resources Institute (WRI) article shows how AI is already helping accelerate the clean energy transition by improving the accuracy of solar and wind forecasts. It allows utilities to better integrate renewable energy into power grids and reduce reliance on fossil fuels. But the same article also revealed a harder truth. AI itself consumes enormous amounts of energy. A typical AI-focused data center can use as much electricity as 100,000 households.
This tension is exactly why AI feels so complex. On one hand, it drives efficiency and innovation across energy systems. Research shows AI-powered neural networks analyze massive streams of data from smart grids. It helps electricity systems instantly adjust supply and demand so energy is not wasted. In more personal spaces, AI shows up in smart meters and home automation apps that break down our electricity use into simple insights. It shows when we consume the most, which devices drain the most power, and how small behavior changes can reduce both emissions and bills.
On the other hand, AI raises concerns about higher emissions and job displacement. A study published in Harvard Business Review estimated that training a model like GPT-4 generated around 300 tonnes of carbon emissions. At the same time, AI is quietly reshaping the types of jobs available to our generation. According to the International Energy Agency, roles focused on administrative tasks (such as, procurement, compliance checks, documentation) are among the most exposed to automation.
And many of us stand in the middle: hopeful, but cautious. Almost every day, I open conversational AI tools like ChatGPT and Google’s Gemini to support my work. For instance, in my current research on green job projections, I often upload raw datasets into these tools to explore patterns of job creation and loss in the energy sector. In seconds, they do what would have taken me days, such as identifying patterns and helping me see connections I had missed. It feels empowering to have a tool that can amplify my impact far beyond my individual capacity. But at the same time, I can’t ignore the questions in the back of my mind: What invisible costs make this speed possible? Who holds my data? And, how will they use it?
Meanwhile, the climate crisis is no longer a distant warning. It is unfolding in real time. Record-breaking heatwaves, floods, and wildfires are reshaping lives and ecosystems across the world. In this reality, awareness alone is no longer enough.
AI did not arrive overnight, it grew up slowly, just like we did. What began decades ago as simple experiments in pattern recognition and rule-based systems gradually evolved into machines that could “see” satellite images and “hear” sounds from nature.
Recent research shows how AI combined with IoT (the Internet of Things) and remote sensing is transforming how we protect forests. Instead of waiting months for field reports, these technologies can monitor forests in real time, detecting illegal logging, plant diseases, and even early signs of wildfires before the damage spreads.
At the same time, AI is learning to recognize birdcalls hidden within thousands of hours of audio recordings. In New Zealand, this allows scientists to identify where endemic birds still live and estimate their populations without disturbing their habitats.
For a long time, AI lived mostly in laboratories, research papers, and expert circles, quietly supporting science, mapping forests, identifying species, and modeling complex systems behind the scenes. Then came a turning point. With the rise of deep learning, transformers, and finally the public release of generative AI tools, AI moved from the background into everyday life. Suddenly, it was no longer just something scientists had used since the 1950s. It was something students, activists, and young people everywhere could interact with directly.
So we are left with a defining question for our generation: if AI is already shaping how we learn, work, and communicate, why shouldn’t it also shape how we respond to the climate crisis?
For our generation, climate action is no longer just about protests or policy papers. It’s also about data, prediction, and decisions made in real time. And this is where AI quietly, but powerfully steps in.
On the mitigation side, AI helps us understand the planet faster than ever before. AI doesn’t just guess, but learns within scientific rules. Instead of relying only on slow, complex climate models, scientists now combine physical climate knowledge with machine learning. This means more accurate weather forecasts, clearer climate scenarios, and better predictions of how biodiversity and ecosystems might change.
AI is also changing how communities respond to climate change on the ground. In Arctic regions, AI tools have been developed together with indigenous communities, blending local knowledge with satellite data to help fishers adapt to shifting ecosystems. In Sanikiluaq, an Inuit community in Nunavut, Canada, AI is being developed not to replace local knowledge, but to strengthen it. An AI tool created by PolArctic combines indigenous knowledge with satellite data and western science to identify the most reliable inshore fishing locations as ecosystems shift due to climate change. This shows something powerful that AI doesn’t have to erase local wisdom, it can amplify it. For young people who care about climate justice, this opens space for technology that respects culture, place, and lived experience.
At a global level, AI helps make climate policy more accessible. Tools that analyze thousands of climate documents across different languages allow policymakers and civil society to see what governments are actually committing to. One example is Climind, an AI platform that uses large language models (LLMs) and retrieval-augmented generation (RAG) to analyze complex climate and regulatory documents. Instead of spending weeks reading technical reports, Climind provides expert search capabilities through an indexed search. For young people advocating for accountability, AI becomes a way to read between the lines of climate promises.
On the adaptation side, AI helps us prepare instead of just reacting. Early warning systems powered by AI now predict floods, droughts, hurricanes, and wildfires using data from satellites, weather stations, and sensors. For example, Siddhartha Patel Daswani, a 17-year-old Indian-American student, founded SmokeSignal, a youth-led initiative that builds AI-based tools to detect and predict wildfires in real time. It gives communities early warnings to protect lives and ecosystems and turns personal experience of wildfire loss into a climate tech platform. For young people working in disaster response, this means technology can save lives, not just optimize systems.
Even though AI feels powerful, it is not neutral, cheap, or automatically fair. One of the biggest risks is that AI itself consumes energy, especially through data centres that often run on fossil fuels. For a generation fighting a climate crisis, it’s uncomfortable to realize that the tools meant to help could also deepen the problem if deployed carelessly.
Beyond emissions, there’s the issue of who gets access. Many young people (especially in Least Developed Countries and Small Island Developing States) face unreliable electricity, limited internet, and weak digital infrastructure making AI-driven climate solutions feel distant or inaccessible. This digital divide risks turning AI into a privilege rather than a public good.
Data security also matters especially for climate-vulnerable regions with weak cybersecurity systems. If people don’t trust how their data is used, they won’t trust AI-driven climate action. But trust is not only about security, it is also about fairness. AI systems trained on incomplete or unequal data can ignore vulnerable communities or reinforce existing inequalities. For example, when climate datasets under-represent certain regions, AI models can produce weaker predictions, leaving already vulnerable communities less prepared for disasters. It risks reinforcing the same inequalities we are trying to solve.
Lastly, youth misrepresentation is especially concerning in many Least Developed Countries, where young people make up a large share of the population. Yet despite this, they are often excluded from decision-making spaces. When young people are treated only as “users” and not decision-makers, AI solutions risk missing local knowledge, lived experience, and grassroots innovation. If AI is going to truly serve climate solutions, young people must move from being users to becoming co-designers, watchdogs, and innovators.
Young people also can help address the risks of AI in climate action by supporting community-based AI initiatives. It means local knowledge (especially indigenous and grassroots perspectives) is integrated ethically and respectfully into climate modeling and adaptation strategies. This shift is already happening.
A project led by the Center for International Forestry Research and World Agroforestry (CIFOR-ICRAF) works with First Nations communities in Canada to integrate indigenous expertise into AI-based fire monitoring and governance tools. It ensures local insight shapes how AI tracks and responds to landscape fires.
Beyond community integration, our generation also has a responsibility to rethink how AI itself is built and powered. One of the biggest opportunities lies in advocating for green AI. AI that runs on renewable energy, uses efficient algorithms, and considers its environmental footprint from design to deployment. Young engineers, researchers, and climate entrepreneurs can prioritize energy-efficient coding, low-carbon data centers, and transparent life-cycle assessments.
However, responsible AI is not only about better design, it is also about skills. The future requires capacity-building. Youth-led innovation hubs, interdisciplinary research between climate science and computer science, and start-ups focused on localized AI solutions can redefine what climate action looks like in developing countries. Instead of importing solutions, young people can build tools that reflect their own realities.
One example of this in action comes from Viet Nam’s Youth Digital Citizen Challenge (AI for Climate Action). This innovation platform engages young innovators across Viet Nam to build AI-driven solutions for climate resilience in the Mekong Delta. Participants developed technologies such as AI tools for salinity forecasting and smart disease detection in aquaculture.
As a Gen Z who grew up in a digital world, I’ve always been told that technology can make the world better. Few things have driven progress as consistently. But history also reminds us that technology, when deployed without intention or accountability, can deepen inequality and environmental harm. The difference is never the technology itself. It is the choices we make around it.
For young people, this moment matters. We are not just users of AI, we are future policymakers, engineers, activists, and community leaders. Whether AI becomes a climate solution or another layer of the problem depends on how boldly and responsibly we act today.
AI is not the hero. AI is not the villain. We decide the story it tells.
Written by Dwi Tamara