Resources

Policy Updates

UW Medicine’s AI Policy

UW Medicine Use of Artificial Intelligence (AI) Policy is part of UW Medicine’s system-wide governance approach to AI and reflects our commitment to safe and appropriate AI use across clinical, research, educational, and operational settings.

Originally established September 2025, the policy outlines when AI tools may be used without review, when uses with elevated risk factors require internal review, and the responsibilities users have for ensuring privacy, security, and accuracy. Recognizing the rapid evolution of AI technologies, regulations and best practices, the policy is intended to evolve over time and will be updated regularly to address emerging risks and opportunities.

Read the AI Policy & the AI Glossary

Policy Updates

Effective September 1, 2026, the policy expanded to include education activities and a broader scope for research. Specifically, the following updates were made:

Policy title change

The policy title has been updated from “Use of Artificial Intelligence (AI) in the Healthcare Setting” to “UW Medicine Use of Artificial Intelligence (AI).” This change broadens the scope of the policy beyond the healthcare setting and provides a simpler, more streamlined title that reflects its overall purpose and applicability across UW Medicine.

Expanded research scope

Research is governed by the policy if it is led by any UW principal investigator (rather than just School of Medicine principal investigators), uses AI, requires UW Institutional Review Board (IRB) review, and utilizes either UW Medicine data or targeted enrollments of UW Medicine patients, staff, faculty or learners (rather than just patients). 

Inclusion of education activities

In addition to covering clinical care, research and UW Medicine business operations, UW Medicine’s AI policy now applies to use of AI within the learning environment, including the clinical learning environment. It also addresses specific requirements for learner use of AI across all UW School of Medicine programs, including WWAMI.

  • Education-specific risk and internal review: The policy includes new characteristics of elevated AI risk (including FERPA, candidate selection) and incorporates a new review workgroup focused on learners and the learning environment.
  • Authorization:  Learners, including trainees, must also receive authorization from their program, supervisor or instructor to use AI in their coursework, assessments, exams and in the clinical learning environment (e.g., clinical documentation, patient care activities, case presentations, in-training or board exam preparation, teaching materials), regardless of whether the tool has been approved by UW Medicine or is considered low risk per the policy.
  • Policy glossary: The AI Policy Glossary now includes additional defined terms such as FERPA, Learner, and Learning Environment. The UW Medicine Data definition was also expanded to include education records.

Frequently Asked Questions (FAQ) Updates

The FAQ were expanded to include broader guidance, with new questions addressing the expanded scope of research and education activities. For education, dedicated sections were added for both instructors and learners. The updated FAQ also includes additional guidance in the following areas:

  • Authorization requirements
  • Responsibilities of faculty, instructors and supervisors
  • Instruction and evaluation
  • Standards for use of AI
  • AI user requirements
  • Inappropriate use
  • Suspected violations
  • Professional error
  • AI literacy and training

AI Intake Updates

Previously, people who didn’t have an AMC account would have to go through a different pathway to submit AI use cases for internal review. Now, we have simplified the intake process. Please note that not all uses of AI require internal review. Review the AI policy and FAQ to determine whether internal review is needed for your specific use case.

Questions

If you have questions that are not answered in the FAQ, please contact the UW Medicine Innovation Core at UWMInnovationCore@uw.edu.