Event Recap: Communicating Course Expectations for AI Use

[dlaufenberg]

On July 8, 2026, the Google for Education higher education community gathered for a critical conversation on navigating the complexities of generative AI (genAI) at the course level. Moderated by Kristina Ishmael, the panel featured Dr. Jon Becker, an associate professor of educational leadership at Virginia Commonwealth University, and Dr. Lance Eaton, the senior associate director of AI in teaching and learning at Northeastern University.

The primary takeaway from the discussion was clear: sweeping, university-wide AI mandates often fail to address the nuanced realities of individual academic disciplines. Instead, educators must prioritize course-level guidelines, fostering collaborative environments where students are actively involved in establishing ethical norms and internal standards for AI use.

The Need for Nuance in Policy

Institutions are grappling with how to regulate a technology that fundamentally shifts the academic landscape. Becker noted that blanket policies struggle because the needs of an English literature course differ vastly from those of a computer science program. Relying purely on detection or prohibition is a losing battle; instead, educators need policies that align with their specific pedagogical goals and the realities of their adult learners.

"I don't think we can actually develop university wide policy on this... what works for an English literature professor is not going to work for an accounting professor or a computer science professor."

Collaborative Norms and Student Agency

A theme throughout the discussion was the importance of centering student voices. Eaton advocated for moving away from unilateral edicts, proposing instead that educators begin the semester by co-creating expectations. By allowing students to annotate a shared syllabus and articulate how they expect to be held accountable, faculty can build trust and secure early buy-in. This process views students not as potential cheaters to be monitored, but as partners in defining productive learning struggles.

"Students are able to make recommendations and articulate what should be the expectations of one another and what should be the expectations of the instructor."

Shifting the Paradigm: From "Can" to "Should"

As the group analyzed the anxiety students face when navigating a patchwork of different course policies, the conversation shifted toward the language we use. Rather than focusing on what AI technically can do, the panelists suggested guiding students to ask what they should do to preserve their own educational investment. Fostering this critical discernment helps students choose to deploy AI as a partner for mundane tasks, rather than shortchanging their own learning.

"I think there might be an interesting reframe here: if the students are asking what can I do with it and what can't I do with it…maybe they should be asking what should I use it for and what should I not use it for..."  Jonathan D. Becker, J.D., Ph.D.

Practical Takeaways for Faculty

• Co-create policies: Begin the term with a collaborative document where students help define acceptable AI use and articulate their own accountability measures.
• Normalize the conversation: Do not rely on the syllabus alone; discuss AI expectations openly and repeatedly before assignments to build trust and reduce student confusion.
• Provide concrete examples: Give students explicit models of appropriate AI use, such as prompting an AI to act as an interviewer for brainstorming rather than simply asking it to generate ideas.
• Reflect on the should: Frame discussions around the ethical and productive deployment of tools, helping students understand the difference between assistance and cognitive offloading.

What's Next?

• Join the Next Discussion: Register for our upcoming July 22nd panel focusing on Academic Integrity in the AI Age.
• Apply for the Current and Upcoming Cohorts: Join our July, August or September Assignment Redesign Cohorts designed to help educators evaluate and strengthen their syllabi through new pedagogical frameworks. Space is limited to ensure deep collaboration.
• Fall Programming: Look forward to upcoming fall webinars specifically tailored for classroom practitioners and their daily instructional experiences.

Shared Event Resources

• Meeting Transcript
• VCU Gen AI Guidelines
• UC Berkeley Law School AI Policy
• EDUCAUSE Framework for Transparent GenAI Use
• Crowdsourced Syllabi Policies for Generative AI Repository

This content was created by a human and refined by Gemini.