Connecticut College AI and the Liberal Arts Symposium

The conversation about AI in higher education is loud, anxious, and often dominated by the largest institutions with the most resources to navigate it. Small liberal arts colleges are watching from a complicated position: wary of the hype (if not downright resistant), uncertain about the risks, and not entirely sure the opportunity being described is one that was designed with them in mind. This presentation will argue that, for small colleges with distinctive curricula, signature programs, and their own institutional culture, AI can offers something more interesting than automation or efficiency gains. It offers, for the first time, the ability to build infrastructure that actually reflects what makes them distinctive. Tools that understand your academic model. Dashboards that speak your institutional language. Systems that serve your students, and not a generalized approximation of them.New College of Florida decided to rethink our approach to software built for someone else and start building tools that are genuinely, specifically ours. In a single academic year, a small cross-functional team (IT, Institutional Research, Registrar, Provost's Office, and students) built working dashboards for advising, athlete eligibility, thesis progress, early alert and retention, and faculty workloads, at a fraction of commercial licensing costs.This session is a practitioner's honest account of how we did it, and why the real payoff isn't efficiency. It's institutional identity.

Higher education is entering a new era where AI is reshaping how institutions engage students, support staff, and operate at scale. Yet many campuses are still struggling to determine what true AI readiness actually looks like.This session explores how institutions can move beyond experimentation to build an AI-ready campus grounded in strategy, governance, and practical implementation. Attendees will learn how colleges are using AI to improve student engagement, streamline operations, and expand institutional capacity while maintaining a human-centered student experience.Through real examples and an interactive readiness framework, participants will examine common barriers to adoption, assess their institution’s current level of AI readiness, and identify practical next steps for moving forward. The session will also explore how AI agents and AI-supported workflows are changing expectations around responsiveness, personalization, and operational scalability across higher education.

What can AI do, and what should it do? My approach to AI and Ethics builds a single inquiry from two disciplines that rarely share a room: computer science, where students implement a neural network from first principles without libraries, and serious philosophy, where they reason about how one ought to act. The pairing exposes a tension at the heart of the liberal arts in the age of AI. If a machine ties your shoes, do you ever learn to tie them yourself—or to do your arithmetic, your writing, your deciding? The liberal arts exist to cultivate precisely the capacities AI now offers to outsource: judgment, reflection, moral agency, the power to be more than a child of one’s time. This talk asks whether teaching AI must mean defending those capacities, and how one course might do both at once.

Large language models (LLM) are being used in educational settings to provide guidance, feedback, and policy-related judgments. However, these systems produce inconsistent responses to similar cases, raising concerns about fairness, transparency, and trust. Rather than relying on probabilistic model output, our approach represents cases in terms of salient features and applies principles derived from institutional academic policies to guide decisions. The LLM remains responsible for interpreting cases, while a separate layer using principles ensures similar cases are treated consistently. This approach offers a middle ground between unrestricted generative AI and rule-based systems.  By making the basis for decisions explicit and inspectable, it provides greater transparency, supports accountability,and allows policies to be operationalized in a consistent form.  We illustrate the approach using examples drawn from our college policies concerning student use of AI and discuss the implications for responsible deployment within the liberal arts.

Title:The AI-Native Classroom: Early Lessons from a Carnegie Mellon PilotAbstract:For the past year, debates about AI in higher education have largely focused on student access to tools such as ChatGPT. Yet access to AI and learning in an AI-native environment are not the same thing. This session examines what happened when AI was integrated directly into course materials, learning activities, and student study workflows in a graduate-level Carnegie Mellon course. Drawing on observations, usage data, student feedback, and faculty reflections, the presentation explores how students actually engage with AI when it becomes part of the learning environment itself. Participants will discuss implications for pedagogy, academic integrity, faculty roles, and the future of liberal arts education in an increasingly AI-mediated world.

Artificial intelligence presents liberal arts educators with a profound contradiction: AI can automate portions of research, analysis, and writing while simultaneously creating new opportunities for inquiry, interdisciplinary thinking, and human-centered learning. This interactive session explores how liberal arts classrooms can utilize AI not to replace student thought, but to deepen interpretation, synthesis, ethical reasoning, and creativity. Moving beyond routine comprehension, this presentation demonstrates how educators can leverage AI as an analytical partner to heighten critical thinking, expand explorative skills, and preserve student intellectual agency. Participants will engage with interdisciplinary classroom activities that combine history, sociology, philosophy, geography, economics, and the arts. Special attention will be given to pedagogical models that emphasize critical consciousness, adaptability, and the synthesis of complex data during periods of technological change. Attendees will leave with practical, scalable frameworks that empower students to bridge the gap between the synthetic capabilities of AI and the sacred, original human capacity for insight, interpretation, and meaning-making.

Artificial intelligence has intensified a question many educators were already beginning to ask: What evidence truly demonstrates learning? For decades, higher education has relied on products—papers, exams, presentations, and projects—as proxies for student understanding. Generative AI challenges this assumption by making it increasingly difficult to determine whether a finished product reflects a student's own thinking. While many institutional responses have focused on detection, surveillance, or restrictions, this workshop explores a different possibility: What if AI is not simply creating an assessment problem, but revealing one that has existed all along? Drawing on experiences redesigning undergraduate and graduate courses through ungrading, authentic assessment, and AI-integrated pedagogy, participants will examine how process-centered approaches can make learning more visible while preserving student agency. Through examples of self-assessment, reflective learning narratives, revision cycles, and AI transparency practices, attendees will explore ways to shift attention from evaluating products to cultivating evidence of growth, judgment, and engagement. Participants are encouraged to bring an existing assignment or assessment practice to the workshop. Using a guided redesign framework, they will identify opportunities to incorporate reflection, process documentation, feedback loops, and authentic engagement in ways that center human learning in AI-rich environments. Situated within the liberal arts tradition, this workshop embraces the paradox that AI simultaneously challenges and clarifies the value of education. As machines become increasingly capable of generating polished outputs, the most meaningful aspects of learning may be those that are difficult to automate: curiosity, ethical reasoning, self-awareness, intellectual risk-taking, and the ability to make meaning from experience.

Oral assessments are often proposed as an “AI-resistant” alternative to written assignments that foster cognitive skills that are especially important in an AI-infused world, such as empathy, conversational communication, and situated-decision making. The unique opportunities offered by oral assessments, however, also come with unique equity challenges that require creative and potentially unfamiliar solutions. Students’ lack of practice with the format, anxiety and performance stress, language and cultural barriers, and neurodiversity (including differences in processing time, speech, and social fluency) all constitute hurdles for equitable assessment, particularly for instructors who have never designed an oral examination. In this session, we will explore these risks, identify strategies for designing and implementing assignments that reduce their impact, and apply those strategies to participants’ specific instructional contexts. Participants will leave the session with a clearer understanding of the equity challenges associated with oral assessment, knowledge of concrete techniques for addressing them, and at least one new or reformed assignment that incorporates inclusive oral assessment components ready to be implemented into their existing courses.

Colleges and universities have barely begun to evaluate the risks and rewards of ubiquitous AI use on their campuses. This session introduces a framework for assessing AI readiness and risk in divisions, departments, centers, and programs. This presentation will highlight key research, legal precedents, and strategies for identifying gaps between policy and pedagogy. Using an experiential approach, attendees will have space to reflect on the profound AI-related changes taking place and consider which roles or areas are most exposed to AI-driven change. Organized into six readiness domains, the session aims to help participants develop a practical, data-driven approach that supports student learning and academic integrity while managing the changing landscape of faculty and staff work. Participants will also have the opportunity to complete a brief AI readiness and risk survey, which offers a diagnostic starting point for administrative or department heads, program or staff leaders, and task force members.

"The sleep of liberal arts produces AI." At the opening of the previous symposium, the keynote speaker shrewdly put forward this statement, proposing the liberal arts education to stay awake in the age of AI. At Colby College, the Davis Institute for Artificial Intelligence endeavors to integrate the two, interrogating and experimenting a liberal arts education organically intertwined with AI. In collaboration with academic and administrative departments across campus, Colby has initiated a responsible AI ecosystem for research, learning, and entrepreneurship. We are developing a multitude of AI programming by situating faculty, students, and staff at the center, while piloting privacy preserving and cost-effective technical infrastructure for campus use. During this session, we will share our approach to a responsible AI ecosystem, attempting to carve out a space and time for the audience to think through a practical AI integration in the context of their home institution.

The rapid integration of artificial intelligence into higher education has created an urgent and widespread need for course redesign, yet most institutions lack the time, funding, and staffing to support faculty at scale. This presentation introduces a structured, AI-assisted course redesign protocol developed at the Yale School of Public Health to address this gap. Designed to be completed in a single two-hour sitting, the protocol guides faculty through a ten-step process that audits learning objectives, assesses AI substitution risk at the assessment level, redesigns for process visibility, authentic student thinking, optional AI integration, and generates ready-to-use syllabus language — all without altering the fundamental nature or intent of the course. Early piloting with YSPH faculty has yielded consistently positive responses, with participants reporting renewed confidence and inspiration around assessment redesign. Attendees will leave with a replicable framework they can adapt and deploy at their own institutions.

AI provides an unmatched opportunity for both accelerating and degrading high quality learning. Institutions of higher education must adapt by providing guidance to faculty and students across all departments; integrating AI literacy into the larger curricula rather than only covering it in specific computer science courses. In order to most effectively plan and evaluate these actions, we argue institutions implement a capabilities-based framework that understands faculty and students in terms of their capacities and functions in relation to existing college missions to produce socially and intellectually responsible graduates. The four capabilities—Access, Association, Involvement, and Literacy—give faculty, staff, and students a shared scaffolding and vocabulary for preserving the cognitive work genuine learning requires. We illustrate this recommendation through concrete examples across disciplines and work with the audience to brainstorm how they can apply this framework to their own classrooms and institutions.

As noted at last year’s conference, liberal arts colleges are in a unique and important position to tackle AI and its effect on higher education. A key facet of this position is our ability to encourage critical thinking in our students across a variety of courses, research, and leadership positions on campus. To that end, and as my institutions continues to grapple with its response to generative AI, I formed an AI student council. The AI student council helps develop programming for students, serves as a sounding board for policies and procedures, and helps with communication. The council serves a complement to the AI Community Council, which brings in professionals, alumni, and other friends of the college, into conversation with Eckerd faculty and staff. This presentation talks about why and how to form the student council, setting goals and objectives for the council, and key takeaways.

As artificial intelligence rapidly evolves, liberal arts institutions face a pressing challenge: how do we swiftly update our curricula to meet the demands of students, parents, and employers across nearly all academic domains? Traditional curriculum development can be slow, but the pace of AI requires agility. A solution is an underutilized resource: our students. They bring unique expertise and lived experiences that can drive curricular innovation.This presentation explores three distinct collaborative structures we implemented to partner students with faculty to develop and revise AI-related curricula.  We have successfully used these models to co-create curricular materials across disciplines, including creative writing, economics, sociology, cognitive science, and ethics. By leveraging these partnerships, institutions can design course materials that are deeply engaging to students, while allowing faculty to gain vital insights into how students actually interact with AI. When using this approach, both sides of the partnership benefit.  Faculty evolve their approach to AI, and students' involvement in curriculum development changes how they approach learning.