From the Campus Operations edition of November 9, 2026
UW's IHME and SEIU 925 open bargaining with proposals that employers not rely on generative AI for employment decisions, disclose its use, and not mandate it
University of Washington Human Resources, Labor Relations · Higher education HR and labor relations · United States (Washington)
- Publisher
- UW Labor Relations
- Original publication
- Updated 9 November 2026 (recap of 27 October and 5 November sessions)
- Source retrieved
- 2026-11-10
What happened
UW Labor Relations' recap of the first two IHME and SEIU 925 sessions lists 'AI impact on working conditions' among union priorities. Union proposals: a performance-evaluation article saying employees' use or non-use of AI shall not be used as a metric; a new generative AI article under which the employer would not rely on GenAI for employment-related decisions, would disclose the nature and extent of GenAI use, would give employees an opt-out for religious or ethical reasons without discipline, and may encourage but not mandate use. The employer's initial proposals were housekeeping edits to unrelated articles. The next session is 18 November 2026.
Why it matters
Gives campus HR and administration leaders a concrete, current example of staff-side terms for AI in workplace administration: disclosure, no AI-use metrics, opt-out and no mandates. It is a bargaining position, not an agreement.
Evidence and measured results
Union proposals recorded in the employer's own recap; nothing agreed. The recap does not state the employer's position on the AI articles. Covers one unit in one research institute.
Limitations and uncertainty
Proposals only; nothing is agreed and the employer has not responded on these articles in the recap. Single bargaining unit (research institute staff), so transferability to other campus units is limited. Different from the Melbourne agreement already archived because it is an open negotiation.
Put this evidence to work
Lighthouse Advisory interpretation, grounded in this source. Enriched 2026-11-09; this does not change the original publication date. Each role follows the same checklist so takeaways compare across sources.
Sales
Role takeaway
- Customer problem
- Administrators lack a map of where AI touches employment processes ahead of bargaining demands.
- Stakeholders
- Chief HR officer
- Labor relations director
- CIO
- General counsel
- Discovery questions
- Which HR and workplace systems have AI features enabled?
- Is AI usage telemetry visible to managers?
- Which bargaining units have raised AI proposals?
- Value hypothesis
- An inventory and disclosure approach may reduce surprise at the table; no outcome is evidenced.
- Potential engagement
- A short inventory of AI in employment-related workflows with a draft disclosure approach.
- Unsupported claims
- Final bargaining outcome
- Employer position
- Applicability to other states or units
Pre-sales engineering
Role takeaway
- Fit
- Campuses whose HR, finance or productivity suites embed AI features visible to managers.
- Architecture and integration
- Configuration review of AI features and usage telemetry in HR and productivity systems.
- Prerequisites
- System inventory
- Telemetry access list
- HR policy owner
- Constraints
- Proposals not agreed
- Single unit
- No technical detail in source
- Security
- Confirm AI usage logs are not repurposed for evaluation and that access to them is restricted.
- Proof of value
- Inventory completeness
- Disclosure notice coverage
- Telemetry access audit
Delivery
Role takeaway
- Work
- Inventory AI in employment workflows, draft disclosure notices and restrict use of AI telemetry in evaluations.
- Dependencies
- HR system owners
- Labor relations
- Legal review
- Ownership
- HR labor relations with CIO support.
- Skills and adoption
- Brief managers on what may not be used as a metric.
- Governance checkpoints
- Pre-bargaining inventory review
- Post-agreement implementation check
- Acceptance criteria
- Proposed: inventory covers all HR systems
- Proposed: disclosure notice issued
- Proposed: no AI telemetry in evaluation templates
- Risks
- Terms change at the table
- Inventory misses embedded features
- Mandating use contradicts agreed terms
Implementation considerations
Lighthouse Advisory interpretation across the operating dimensions a public-sector buyer must settle before this evidence becomes a design. Each note answers the question under its heading for this specific source.
Architecture and integration
What must connect, and where does the AI sit in the workflow?
No technical content; implies AI features embedded in staff workflows would need to support disclosure of use and non-mandatory adoption.
Governance
Who approves, reviews and stays accountable for outcomes?
Proposed obligations: disclosure of GenAI nature and extent, no reliance on GenAI for employment decisions, and no AI-use performance metrics. These would require HR to inventory where AI touches employment processes.
Security and privacy
What data, permissions and controls need testing?
Proposal references protection against rogue GenAI but gives no definition; data handling terms are not described.
Accessibility and workforce
Who is affected, and what skills or accommodations follow?
Directly about workforce terms: opt-out rights, no mandates, AI use not an evaluation metric.
Procurement
What should contracts, pricing and exit terms secure?
Contract terms could constrain how HR and managers buy or mandate AI tools for staff.
Operating model
Which teams own the service once it runs?
Campuses that measure staff AI usage or require adoption should expect these to become bargaining topics.
Publication history
- 2026-11-09Campus Operations · Issue 613 resources
Stable resource ID: uw-ihme-seiu-925-genai-bargaining-proposals-20261105