Weigh potential uses in energy and climate work against the resources required to build and run AI systems.
Up to 60 seconds. Preview before submitting.
Record up to 60 seconds. Preview your video before you submit it.
AI is the only tool fast enough to match the speed of climate collapse. Climate change operates on timescales that human analysis cannot keep up with. AI models process satellite imagery, ocean temperature data, and atmospheric readings in real-time across the planet. DeepMind's AI reduced Google data center cooling energy by 40%. Climate AI startups are discovering new battery chemistries 1000x faster than traditional lab work.
Without AI, we are flying blind on the most complex system on Earth. Earth's climate is a chaotic system with trillions of interacting variables. No human team can model it accurately enough for confident policy. AI weather models like GraphCast already outperform traditional forecasting. Refusing to use AI for climate because of its energy cost is choosing ignorance over imperfect action.
The energy cost is temporary — the climate benefit is permanent. Training GPT-4 used significant energy, but a single AI model that optimizes global shipping routes or building HVAC systems saves orders of magnitude more energy than it consumed. Invest carbon now, save carbon forever. Solar and wind are rapidly making AI's energy footprint renewable anyway.
We already know how to stop climate change — we lack political will, not computing power. The solutions exist: stop burning fossil fuels, electrify transport, insulate buildings, protect forests. None require AI. Framing climate as a technology problem rather than a political one lets governments and corporations delay action while waiting for a magic algorithm that will never arrive.
Big Tech is using "climate AI" to justify building more data centers. Microsoft, Google, and Amazon have all missed their carbon neutrality targets — largely because of AI. Microsoft's emissions rose 30% in one year. These companies fund climate AI research as PR cover while their core business demands exponentially more energy. The fox is guarding the henhouse.
AI-driven resource extraction is accelerating the problem, not solving it. AI is used to find new oil deposits, optimize mining operations, and enable faster deforestation through satellite analysis. For every climate-positive AI application, there are ten extractive ones. Without regulation, AI optimizes for profit — and fossil fuel companies have more money than climate nonprofits.
