AI Agent Operational Lift for Continuing Education At The University Of Utah in Salt Lake City, Utah

Why AI matters at this scale

Continuing Education at the University of Utah operates as a mid-sized, self-sustaining unit within a major public research university. With an estimated 201-500 employees and annual revenues likely in the $40-50M range, it sits in a sweet spot for AI adoption: large enough to have meaningful data assets and a diverse program portfolio, yet small enough to pilot new technologies without the paralyzing governance of the full university. The unit delivers non-credit professional certificates, workforce development programs, and personal enrichment courses to thousands of adult learners annually. Its primary business challenge is balancing mission-driven access with the need to generate revenue, all while proving the employment outcomes that justify tuition costs. AI offers a direct path to scaling personalized services, optimizing marketing spend, and creating the adaptive learning experiences that modern professionals expect.

Three concrete AI opportunities with ROI framing

1. Intelligent enrollment and retention engine. The division likely collects vast amounts of data on learner demographics, course browsing behavior, past completions, and engagement patterns. A machine learning model can score each learner's likelihood to enroll in a specific program and their risk of dropping out. By integrating these scores into the CRM and advisor workflows, the unit can trigger personalized nudges—such as a timely email about an upcoming cohort or a call from an advisor. The ROI is direct: a 5% increase in course completion rates translates to higher revenue and stronger outcome metrics for marketing. Retention improvements also reduce the cost of acquiring new learners to fill seats.

2. Labor market-responsive program design. Continuing education units must constantly refresh their catalog to stay relevant. An AI system that ingests real-time job postings, partner company needs, and regional economic data can identify emerging skill gaps before competitors. This allows the curriculum team to prioritize new certificate programs in areas like AI ethics, renewable energy technology, or healthcare informatics. The ROI comes from faster time-to-market for high-demand programs and stronger corporate training contracts. Instead of guessing what employers want, the division can use data to justify investments to university leadership.

3. Generative AI for instructional design and learner support. Faculty and instructional designers spend hundreds of hours creating course outlines, writing quiz questions, and developing case studies. Generative AI tools can produce first drafts of these materials, cutting development time by 30-40%. Additionally, a fine-tuned chatbot trained on the course catalog and policies can handle routine student inquiries 24/7, from "What prerequisites do I need?" to "How do I request a refund?" This frees professional advisors to focus on high-value career coaching. The ROI is measured in staff productivity gains and improved learner satisfaction scores.

Deployment risks specific to this size band

A unit of 201-500 employees faces distinct risks. First, IT resources are likely limited, with a small team managing integrations between the student information system (likely PeopleSoft), the LMS (likely Canvas), and the CRM (likely Salesforce). An AI project that requires deep custom integrations could stall without dedicated engineering support. Second, data governance is a minefield; continuing education serves non-traditional students, but their data still falls under FERPA and university privacy policies. Any predictive model that uses demographic or behavioral data must be audited for bias and fairness. Third, faculty and staff resistance can derail adoption. Instructors may fear that AI-generated content threatens their expertise, while advisors may worry about job displacement. A change management plan that emphasizes augmentation over replacement is critical. Finally, the unit must avoid vendor lock-in with AI startups that may not survive long-term, favoring established platforms or university-system-wide contracts where possible.