On Tuesday, voters in a minimum of 29 U.S. jurisdictions approved measures that either ban or place new restrictions on the development of artificial‑intelligence data centers, according to reporting by Decrypt and The Washington Post. The ballot initiatives, which appeared on November 3 in ten states, reflect a growing backlash against the massive electricity, water and grid‑capacity demands of AI compute facilities.

Geographic spread and vote outcomes

Ohio saw the most decisive results, with all ten charter measures backed by the grassroots group Conserve Ohio passing. More than 75% of voters in Hubbard supported a ban, while Granville, Trenton and Sunbury recorded approval margins between 78% and 82% for restrictions. In the Midwest, Illinois’ Joliet Township rejected an advisory proposal favoring AI data centers by nearly 80%, and Janesville, Wisconsin, saw over 60% of voters demand a further public vote before any major development at a former General Motors site linked to a proposed data center.

Nature of the measures

The ballot items varied in legal force. Some were binding bans that would directly halt construction, while others were non‑binding advisory votes that signal public sentiment without automatically stopping projects. Thirteen measures were framed to support data‑center development; rejecting those also counts as a defeat for developers. Overall, 28 of the 41 measures tracked by Ballotpedia were written so that a “yes” vote would curb development.

Industry response and broader political context

AI firms are already feeling the pressure. Companies such as OpenAI have begun rebranding their facilities as “compute campuses” to distance them from the negative perception of data‑center projects. At the same time, political figures campaigning on stricter oversight have won office, including Democrat Gina Hinojosa in Texas and Ohio governor‑elect Amy Acton, both of whom support moratoriums on new AI infrastructure. BlackRock CEO Larry Fink warned that a nationwide moratorium could trigger a recession, underscoring industry concerns about the economic impact of blanket restrictions.

What to watch next

Future developments will hinge on whether state legislatures translate these local votes into broader policy, how AI companies adjust their siting strategies, and the outcome of pending lawsuits challenging municipal authority over utility infrastructure. Monitoring utility rate changes, grid‑capacity studies, and further ballot measures in the 2027 election cycle will be crucial for assessing the long‑term trajectory of AI compute expansion in the United States.