Source: AI Governance, Ethics and Leadership
The 2026 midterms exposed a sharp disconnect in how Americans engage with AI policy: voters decisively rejected 40+ data center proposals across 13 states through direct ballot measures, yet showed virtually no appetite for addressing AI's actual governance gaps—model safety, deepfakes, job displacement, or algorithmic regulation all went uncontested at the ballot box. Public concern about AI is hyperlocal and infrastructure-focused (noise, land use, energy), not systemic. That leaves the harder regulatory questions to be fought in legislatures where tech lobbies have deeper leverage and community input is minimal. Winning candidates who campaigned on AI "pauses" and tighter rules are now accountable only for blocking shovel-ready projects, not for building coherent AI governance frameworks. That gap will likely push compute infrastructure to friendlier jurisdictions while leaving the industry's core risks unaddressed.