AI Cannot Solve Climate Change Without Public Control Over Its Development
Source: Noah Gordon. "Could AI Fix Climate Change? Probably Not Under Capitalism.." October 7, 2026. jacobin.com
The Gist
The author argues that even if AI turns out to be a powerful tool, it won't actually fix climate change as long as it's developed and controlled by profit-driven corporations, because companies — including fossil fuel firms — will use AI to protect their own interests rather than transition away from fossil fuels. The only way AI could meaningfully help solve the climate crisis is if it were brought under public, democratic control rather than left to tech billionaires and corporate competition.
Conclusion
AI is unlikely to solve the climate crisis unless it is taken under public control, because under capitalism, technological progress does not reliably translate into the political and economic changes needed to avert catastrophic climate outcomes.
Premises
- AI boosters like Jensen Huang make vague, unfalsifiable promises that AI will 'solve' climate change without specifying mechanisms, echoing earlier economic models (e.g., Nordhaus's DICE) that assumed a future 'backstop technology' would emerge to justify inaction now.
- Even if AI produces genuine technological breakthroughs (better batteries, energy storage, weather forecasting), these advances do not automatically change the underlying political and business incentives of a capitalist system.
- Fossil fuel companies have equal access to AI tools and are already using them to improve extraction (e.g., cheaper, more accurate drilling), and a Nature study found AI productivity gains in oil and gas outweigh gains in renewables.
- Owners of fossil fuel assets (pipelines, gas plants, coal mines) have structural incentives to continue operating and profiting from existing infrastructure rather than shutting it down early, regardless of better alternative technologies being available.
- Competitive pressures among AI firms (and between the US and China) mean that even well-intentioned companies cannot afford to slow down or redirect research for climate-safe purposes, since competitors will not.
- The rhetoric of technological uncertainty around AI mirrors historical climate-denial rhetoric, serving to delay concrete action by suggesting an uncertain future techno-fix will arrive in time.
- AI's uncertainty cuts both ways: it could either solve the climate crisis or destroy civilization, and this binary framing is being used to reduce the perceived urgency of pursuing concrete climate action today.
Assumptions
- Capitalist systems inherently prioritize short-term profit and competitive advantage over long-term collective welfare, even when better technologies exist.
- Public or democratic control over AI development would better align technological progress with climate goals than private, profit-driven control.
- Technological capability alone is insufficient to change entrenched economic interests tied to fossil fuel infrastructure.
- AI companies' public statements about their technology's transformative potential are not fully reliable guides to actual outcomes, given their financial self-interest in hype.
- The climate crisis requires near-term, concrete action rather than reliance on speculative future breakthroughs.