Route adaptive AI ethics instruction by splitting non-users from sustained users, using usage frequency first and familiarity as a secondary check¶

Strategy · All strategies
Evidence · no claims cited

Description¶

The article recommends that short pre-instruction surveys use reported usage frequency as the first intake variable for grouping learners and self-rated familiarity as a secondary check, rather than routing by prior coursework or workshops. The authors state that the results "are more consistent with a split between non-users and sustained users than with a split based on prior coursework or workshops," and suggest heavy users may need additional calibration-focused activities on failure modes.

Design Implications¶

Context¶

Requirements¶

• A short intake survey collecting reported usage frequency and self-rated LLM familiarity before instruction begins

Constraints¶

• The study is cross-sectional and associational; whether differentiated instruction based on these profiles improves learning outcomes requires an intervention study

Target Learners¶

• graduate and postdoctoral bioscience trainees in required research ethics courses

Target Learning Goals¶

• AI ethics readiness, trust calibration, and awareness of over-reliance and LLM failure modes

Affordances¶

• Engagement Intensity Intake Construct

Related Strategies¶

Examples¶

-

Key Sources¶

• Yongkyung Oh, Lynn Talton, Alex Bui. (2026). Engagement Intensity as a Learner-Modeling Signal for Adaptive AI Ethics Instruction. KDD 2026 AI for Education Day, at the ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD '26). https://arxiv.org/abs/2606.18548