Early-alert and predictive systems act on students: they score, flag, and route. Generative AI acts with students: it drafts essays, explains concepts, answers advising questions, and simulates conversation. The theory questions shift accordingly, from “who gets flagged” to “what learning and connection does this tool replace or support.”
The central risk is what this resource calls student success bypass: a student meets the metric, such as a submitted assignment, a completed form, or an answered question, without the developmental process the metric was meant to reflect.
- Success bypass
- Outputs stand in for learning. Grades and completion may hold steady while involvement and competence decline.
- Self-efficacy
- Heavy reliance can weaken students’ confidence in their own ability, or build it when tools scaffold rather than substitute.
- Involvement
- Astin’s premise is that students learn in proportion to their investment of time and energy. Tools that remove effort can remove the investment.
- Belonging
- An always-available assistant can lower the cost of asking for help, or quietly replace the human contact that builds connection.
- Academic integrity
- Policies vary course to course. Unclear rules create anxiety and uneven consequences, often for students with less insider knowledge.
AI literacy for faculty, staff, and students
- How generative tools work, where they fail, and how to check their output.
- When use is permitted, expected, or prohibited, stated course by course in plain language.
- How to use AI to extend thinking, such as feedback, practice, and explanation, rather than replace it.
- What happens to data students enter into campus and third-party tools.
In the Campus Enterprise Model
Where this work sits in campus systems, capabilities, and data.