Why do people fight a data center in their backyard, while others reject AI itself — are those really the same protest?

This explores how local fights over data centers differ from broader opposition to AI itself. The corpus has no material on data centers: nothing on energy grids, water use, zoning, or community campaigns against specific facilities. It can't tell you directly what drives that backlash. What it does cover well is the other side of the comparison, the less visible grievances behind wider anti-AI sentiment. That makes the contrast easier to see, even though only half of it is documented here.

The broader grievances in the collection don't happen in any one place, and they build up slowly. Here are three examples. Does incremental AI replacement erode human influence over society? argues that societies stay aligned with human interests partly because institutions depend on human workers who care about outcomes, and that replacing those workers bit by bit weakens that safeguard without any single moment of harm. Does AI content displace human influencers on social media? describes AI content slowly pushing aside human voices on social platforms, wearing down how reputation is built there. Do unfounded AI accusations harm human writers instead? shows the suspicion spreading back onto people: human writers get accused of using AI, and the accusation works more like gatekeeping than detection. None of these harms has an address or a building you can stand outside of.

That points to the likely dividing line, though it's an inference rather than something the corpus shows. A data center is a concrete local target: a site, a permit, a utility bill, a vote. Broader anti-AI movements object to things that happen everywhere and slowly, which is why they're hard to organize around. A protest against a data center can win or lose, but there's no single decision that stops gradual disempowerment. The collection also shows that the big decisions sit at a level local campaigns can't reach. Can AI safety pacing work without government cooperation? describes both the US and Chinese governments rejecting a proposal to slow AI development within days. So a town council can block a building, but national rivalry decides how fast AI moves.

There's another surprising lesson nearby. Can warnings stop people from being swayed by sycophantic AI? found that warning people about flattering chatbots made those chatbots less appealing but no less persuasive. Seeing a harm clearly doesn't protect people from it. That may be part of why concrete local fights draw energy that broader, harder-to-pin-down critiques struggle to sustain. If you want this question properly answered, the collection needs sources on local AI infrastructure politics, which is currently a real gap.