In April, we had the opportunity to attend the 2026 AAC&U Conference on Learning and Student Success, or CLASS, in Tucson, Arizona.

This year’s conference centered around the theme of courageous care, with sessions exploring topics from campus culture and community building to the current state of AI literacy initiatives for students and staff. During the conference, we facilitated a roundtable discussion on integrating undergraduate teaching with generative AI, sharing insights from Ithaka S+R projects exploring AI in higher education and collaborative AI literacy cohorts.

Below, we each share our personal reflections on the conference and present our takeaways on emerging themes.

Keeping the person in personalization (Michael Fried)

At the recent AAC&U Conference on Learning and Student Success, my colleagues and I facilitated a conversation on integrating undergraduate teaching with generative AI. This roundtable discussion was inspired, in part, by our work with centers for teaching and learning across the California State University system. The lively and wide-ranging conversation included a fair amount of lamentations about AI’s potential to negatively impact teaching and learning. However, several more optimistic themes emerged in which the possible benefits of AI shone through.

Creating personalized learning materials is one such area of possibility. Educational technology circles have long aimed to personalize learning and improve student outcomes by leveraging big data, adaptable learning environments, and evidence-based teaching practices. In the context of this conversation about AI and teaching, the balance shifted: generative AI’s ability to create almost instantaneous, bespoke learning materials takes personalization further than previously imagined.

Attendees at our session agreed that this level of AI customization also creates an opportunity for instructors to center the “person” in personalization, and continue to keep humans in the loop when implementing AI into organizational and educational systems. For decades, research has shown that the instructor-student relationship is one of the most important elements of the teaching and learning process, and it remains a strong predictor of positive student outcomes. Leveraging AI could free up instructor time and attention to build relationships with students.

Although generative AI remains a disruptive force in teaching and learning, the participants in our discussion agreed that relying on fundamental principles is one way to mitigate some of AI’s negative impacts. Maintaining, or establishing, the college classroom as both relational and developmental for instructors and students will be essential to successfully integrating AI into teaching and learning. Participants agreed that the time AI saves instructors should be reinvested in fostering connections with students rather than simply presenting more content.

AI literacy across disciplines (Nathan Kelber)

At CLASS, AI in the classroom was front and center. The conference brought together faculty from across disciplines to grapple with what has become one of the most pressing challenges in higher education: how to integrate AI literacy into teaching and learning. Presenters opened with what has become a familiar refrain—“Since ChatGPT released in November 2022…”—but the conversations that followed revealed how differently disciplines are experiencing this moment, and how much common ground they share.

What emerged across sessions was a pattern that resonates with findings from Ithaka S+R’s Integrating AI Literacy into the Curricula cohort project: AI literacy development is happening from the ground up, driven by motivated individuals and units, often without coordinated institutional strategy. CLASS offered a vivid snapshot of where that ground-level work stands—and what is getting in the way.

Faculty exhaustion as an institutional condition

Some disciplinary divides played out predictably, with a handful of English and history faculty calling for a return to blue books and handwritten exams. It would be easy to dismiss these responses as resistance to change, but they reflect something more systemic. Faculty who voiced skepticism about AI also spoke of pandemic recovery fatigue and ongoing budget cuts. Regardless of whether their critiques were political, philosophical, or ecological, the common thread was exhaustion—both mental and emotional.

STEM faculty expressed parallel concerns. Computer science students nearing graduation questioned the long-term value of learning to code, a skill that has been the gold standard in technical employment. Science faculty noted that students are struggling with foundational skills, particularly the ability to evaluate when LLM outputs are unreliable or simply wrong. These concerns cut across the disciplinary divide and point to a shared challenge: faculty across fields are being asked to adapt their teaching at a moment when institutional capacity is already strained.

Shame as a barrier to faculty engagement

One of the most compelling sessions, “Deshaming AI in Education,” offered a framework for understanding why faculty engagement with AI has been uneven. Jason Blomquist and Kelly Arispe of Boise State University identified two different dimensions of faculty shame, one that is: internal (“I am not an expert and I am falling behind”), and one that is social (“My colleagues will think I am lazy or that I am betraying my craft”). To counter, Blomquist and Arispe provide “tactical de-shaming” by creating low-stakes environments that put humanistic concerns front and center.

The case for interdisciplinary collaboration

Speakers from California State University Maritime Academy provided an instructive example of taking a cross-disciplinary approach to AI literacy. Taiyo Inoue and Ariel Setnike (math) have partnered with Sarah Senk and Aparna Sinha (English) to address AI literacy gaps in gateway courses. The Cal Poly Maritime partnership highlights that the humanities have a distinctive contribution to make. The most valued skill in what some have called the judgment economy may be rhetorical—the ability to evaluate, marshal, and critique an endless stream of AI-generated content. Composition, argumentation, and critical reading have always been cross-curricular concerns; AI makes them urgent ones. But advancing AI literacy still requires the kind of big tent approach that CLASS modeled, where disciplinary perspectives inform rather than fragment the effort.

Looking ahead

The CLASS conference confirmed that higher education has moved past initial reactions to generative AI and into a more substantive phase of engagement. Faculty across disciplines are developing AI literacy programming, experimenting in their classrooms, and confronting the institutional barriers—exhaustion, shame, fragmentation—that slow this work down.

The challenge now is coordination. As we’ve found through our cohorts, grassroots efforts need institutional support to become sustainable. Leadership has a role to play in providing resources, strategic vision, and the cross-campus coordination that individual faculty and units cannot achieve on their own. The conversations at CLASS suggest that the will is there. The question is whether institutions can match it with the structures and support to make AI literacy a shared, lasting priority.

Care at every level (Zhuo Chen)

The theme of this year’s CLASS Conference was courageous care. To care, in higher education, is to tend intentionally to the wellbeing of the whole community: leaders who build supportive cultures, faculty and staff who show up for each other, and students whose psychological wellbeing and intellectual development are cultivated with equal seriousness.

Kevin McClure’s plenary talk urged higher education leaders to prioritize faculty and staff wellbeing, arguing that a caring workplace will create downstream benefits for student success. He sees the faculty and staff experience as a precursor to everything that takes place in the classroom. Staff working conditions directly influence student learning conditions. Burned-out faculty and staff will not be able to deliver the experience students need, and talent retention may also decline as a result. He shared an example from Miami University in Ohio: after the marketing department lost a significant number of staff, an employee culture survey revealed that a leading cause of turnover was a lack of room to grow, especially for staff who were not interested in being supervisors. Strategies like designing different tracks of career pathways, one for staff who wanted to pursue a supervisor role and one for staff who didn’t, have proven to be successful in addressing this issue. McClure also highlighted another example of intentional care at the University of Louisville, where dedicated staff support the employee experience by providing onboarding and training.

Jennifer Brady and Rebecca Ropers from University of Minnesota Duluth shared initiatives aimed at building a caring and supporting campus culture for all. Through the Hub for Integrated Learning and Leadership (the HILL), they adopted a collegiate strategy to provide professional development opportunities for staff and faculty. Low cost affinity and book groups turned out to be very effective in community building, creating bridges over the silos to surface common themes of interest. Institutional leadership also formed a learning community to strengthen connections through reading groups and workshops.

Building a supportive and caring campus culture is also key to enhancing student success. In addition to cultivating students’ intellectual needs, faculty should make efforts to foster students’ psychological safety and wellbeing. AI—a big topic of the conference—seems to pull these two caring obligations in different directions. Kathleen Kennedy shared her campus-wide study that shows AI can be used to narrow the gap in academic performance between different student populations. For first-generation college students in particular, using AI responsibly for academic study could help alleviate the psychological stress caused by the academic gap. At the same time, higher education has been searching for ways to mitigate the potential harm of AI usage on students’ critical thinking skills. In this way, higher education leaders and staff are navigating the benefits and tradeoffs of AI—hoping to harness its benefits while guarding against its harmful effects—as they build supportive care for the students.