Is AI preparation still something graduate programs can treat as optional? (slide text and speaker notes)

Is AI preparation still something graduate programs can treat as optional? — slide text and speaker notes _Mohammed A. A. Abulela, MetaMetrics, Inc. / University of Minnesota (with co-authors Guher Gorgun, Brian French, Brian Leventhal, Matthew Gushta). AIME-Con 2026, 7 October 2026. **Transcribed verbatim from the slide images** of `NCME AI Panel.pptx` (the slide text lives in images, so the file has almost no machine-readable text), followed by the presenter's speaker notes. Citable in public._ Slide 1 — One Claim Slide title: Is AI preparation still something graduate programs can treat as optional? **One Claim.** AI-related preparation should no longer be treated only as an optional extension of graduate training in educational measurement and psychometrics. It should become an intentional part of professional preparation, while remaining grounded in validity, fairness, measurement theory, and psychometrics. Slide 2 — Evidence: One Concrete Example Slide title: Evidence. Subtitle: A clearer picture of preparation and demand. Side caption: Bridging Academic Preparation and Workforce Expectations. **Doctoral programs — limited inclusion of computational or AI coursework.** Across 90 doctoral programs, qualifying computational or AI-related coursework was identified in only 16 programs (17.8%). [Bar: 17.8% — 16 programs out of 90.] **Job advertisements — strong demand for computational or AI skills.** Across 44 measurement-related job advertisements, 79.5% included at least one computational or AI-related area. [Bar: 79.5% — 35 job ads out of 44.] **Speaker notes.** Most coursework was in stats programs (Artificial Intelligence, Machine Learning, Natural Language Processing, Data Science, Data Mining, Programming/Computational Skills) Jobs: - GenAI/LLMs - Prompt Design/Fine-Tuning - AI-Enabled Automated Scoring - AI-Enabled Automated Item Generation - AI Model Evaluation/Validation - Human-Machine Agreement - AI Fairness/Bias/Responsible AI

Most coursework was in stats programs (Artificial Intelligence, Machine Learning, Natural Language Processing, Data Science, Data Mining, Programming/Computational Skills) Jobs: - GenAI/LLMs - Prompt Design/Fine-Tuning - AI-Enabled Automated Scoring - AI-Enabled Automated Item Generation - AI Model Evaluation/Validation - Human-Machine Agreement - AI Fairness/Bias/Responsible AI Slide 3 — One Unresolved Professional Problem Slide title: Unresolved Issue. **What is the minimum AI-related preparation that every future measurement professional should have?** What should be required? What can remain elective? How do we integrate AI without displacing validity, fairness, and psychometric foundations? **Speaker notes.** We do not yet have a disciplinary answer. Should every graduate understand how to evaluate an AI-enabled scoring system? Should they be able to examine bias in an LLM-based assessment? Should they understand prompt design? Machine learning? Human-AI agreement? And which of these belong in required preparation rather than electives? Then connect it directly to the NCME framework: The professional problem is not simply deciding which new AI tools to teach. It is determining which AI capabilities are becoming part of competent measurement practice and how to teach them without displacing the validity, fairness, psychometric, and measurement foundations that define the field. Slide 4 Thank You. Email: mabulela@metametrics.com · mhady001@umn.edu