Yale’s staff union seeks five-year job guarantee as AI enters contract talks

Yale’s clerical and technical workers want more than assurances that AI will help them work: they want a staffing guarantee in their next contract. Local 34 is seeking a five-year floor of 3,933 positions as it challenges the university’s AI rollout. Yale denies cutting positions in favor of AI and says the tools support staff.

The demands are part of ongoing contract negotiations, detailed by the New Haven Independent on October 28. They follow a report Local 34 released in early October criticizing Yale for introducing AI-driven changes to administrative work without union input.

The concern is vacancies, not just layoffs

The union’s report describes changes across grant writing, memos, email outreach, customer service and information retrieval. Its central allegation is what it calls “passive replacement”: using AI to take over or redistribute work after employees quit or retire, rather than replacing those workers with new hires. Yale disputes that account.

Local 34 points to a hiring freeze and retirement incentive as staffing policies that could make that possible. In the union’s account, a smaller workforce supplemented by AI could keep doing the same volume of work while the university avoids filling vacancies. The concern therefore extends beyond whether an employee is directly dismissed because of a chatbot.

The report puts Local 34’s membership at about 3,933, down from 4,112 in April 2025. Those are union-reported membership figures, not a count of jobs shown to have been lost to AI. At the time of the Independent’s article, Yale also had public postings for at least 17 administrative assistant positions.

Yale has not reduced positions in favor of AI tools but instead is focused on ways to make more time for staff to do human-centered work or to increase external access to Yale’s resources

A Yale spokesperson, in an email to the New Haven Independent

The tools behind the dispute

Yale began rolling out its internal AI platform, Clarity, in August 2024. It offers access to versions of Claude, ChatGPT and Gemini designed to meet certain security standards. Yale’s website says eligible users can also create and manage custom AI agents at no cost. Departments have put chatbots to work on library searches, registrar information and finance questions.

Employees’ experiences in the union report differ sharply. Administrative assistant Carolyn Kyrtopolous said generative AI reduced the time needed to produce scholar biographies by three quarters. Another assistant, Jennifer Ongley, described erroneous output that took extra time to repair. These are individual workers’ accounts, not a university-wide measure of productivity.

Dana Greene, who retired in May after 32 years at Yale’s School of Public Health, told the Independent that supervisors repeatedly encouraged Clarity use before her departure. At team meetings, she recalled being told to ask a chatbot questions before approaching a manager. For travel questions, it cited the same university policies she would otherwise have consulted herself.

The school’s Helen Helpwell chatbot illustrates the current boundary. It answers research-administration questions about matters such as budgets, hiring and policies. But it cannot access Workday, Yale’s billing and personnel platform, or book a flight. Greene’s former job also involved coordinating events, reconciling credit card expenses and helping faculty with course logistics.

From assurances to contract terms

Beyond the staffing floor, Local 34 is seeking several protections in the AI transition:

• A union role in shaping the transition to AI-assisted work.
• Protection for existing jobs and stronger layoff benefits.
• Higher salaries for employees trained to incorporate AI into their work.

Yale says individual schools and teams have room to implement AI according to their needs. Its spokesperson described strong interest in training and argued that making routine tasks more efficient could free employees for more personally fulfilling, human-centered work.

Local 34’s report says the rollout has made negotiations more challenging. Its proposed staffing commitment is intended to give workers confidence that AI will improve university operations without eliminating jobs. That commitment remains a union demand in the contract talks, rather than an agreed guarantee.