Oregon moves first with strict AI rules for state government amid federal gridlock

On September 23, Oregon Governor Tina Kotek signed a new executive order that sets out “responsible AI procurement standards” for state agencies. The order aims to tighten safeguards around how Oregon buys and deploys AI systems, with particular emphasis on powerful frontier‑scale generative models.

Under the directive, any AI system used by state government must undergo upfront assessments for risks such as bias and discrimination, privacy violations, and security vulnerabilities. It also codifies basic AI governance principles — including requirements for human oversight and expectations that systems be explainable enough for officials to understand how they reach decisions.

Kotek framed the move as a response to weak federal action, arguing that states have to step in to protect residents’ rights and safety. Oregon’s approach overlaps with efforts in states like California, but it stands out for using public procurement as the main lever to enforce standards. For governments elsewhere, including local authorities in Japan, the order offers a concrete template for how to put responsible‑AI ideas into practice inside everyday public services.