Solarpunk and the AI era

Most visions of an AI future collapse into two camps: techno-utopia where friction disappears, or surveillance dystopia where everything does. Both are useless for building. Solarpunk — a movement born in 2008 as a rebellion against the relentless dystopias of science fiction — offers the third vision: a sustainable future interconnected with nature and community, grounded in the world we actually have. The "solar" is renewable energy and the refusal of climate doomerism; the "punk" is the DIY ethos of building that future without waiting for permission.

The timing is why this matters now. We're living climate change rather than predicting it, while simultaneously building the most powerful problem-solving tools in history. Tools inherit the values of their creators and the systems they operate within. AI built inside the extractive, growth-at-all-costs paradigm that produced the climate crisis simply accelerates toward the same cliff. AI built on different principles might produce something worth inheriting.

Those principles have teeth, and they read like an antidote to most of modern tech's pathologies: decentralization over corporate control, repair over replacement, community over individualism, and appropriate technology — progress redefined from maximizing profit to optimizing the intertwined health of people and ecosystems. Solarpunk explicitly rejects greenwashing; plants on a luxury condo aren't the point. Technology is scaffolding for living systems, not a controlling force — which makes choices like open weights, edge computing, and models that run on modest local hardware political decisions, not just technical ones. And because we optimize what we measure, measuring only inference speed and accuracy gets us fast, precise, extractive systems; measuring energy, accessibility, and repairability gets us different ones.

For us this isn't aesthetics — it's how we run engineering. Choosing open source and contributing back. Optimizing for full lifecycle cost, not nominal benchmarks. Building for maintainability and accessibility from the start. Being transparent about limits and trade-offs. And asking "should we build this?" in the same breath as "can we."

The AI revolution is happening regardless. The question is never whether we'll get an AI-powered future — it's what kind. Optimism, done seriously, is an act of rebellion: imagining a better future is the first step to building one, and building it anyway — even when the incentives push the other way — is the punk part.