Is This Moment Actually New?

At this point, it is worth asking whether this moment is truly new or whether it is another iteration of a familiar pattern. Higher education has already experienced several technological shifts that prompted similar concerns. Calculators raised questions about the loss of mathematical skill. Search engines reshaped how research was conducted. Word processors changed the writing process itself. Socrates warned that the act of writing things down threatened thinking. In each case, there were warnings about overreliance and diminished ability. Those warnings were not entirely wrong, but they also did not fully capture what was happening. The tool changed, but more importantly, the relationship to the task changed.

What changes with new tools is rarely just the task. It is the relationship to the task.

What Changes When the Tool Responds?

Artificial intelligence appears to follow this pattern, but it also introduces a complication that earlier technologies did not. A calculator does not respond to uncertainty. A search engine does not refine your question in real time. Generative AI does both. Students do not simply input a query and receive an answer. They revise their prompts, adjust their language, and ask again. There is a back and forth that begins to shape how they approach the task.

This raises a question that is not yet fully addressed in most higher education conversations. At what point does repeated interaction begin to alter how a student experiences the learning process? My research on human-AI companionship examines this more directly by asking what kind of relationship a system invites: helper, assistant, guide, or something else entirely. Students may not articulate this in theoretical terms, but they respond to it in practice. They return to the same systems, develop preferences, and begin to rely on them in moments of uncertainty. The system does not possess intention or awareness, but the interaction still produces a relational effect.

If interaction is beginning to shape how students engage with their work, then it also begins to shape how we should think about evaluating that work.

The system may not be relational, but the experience of using it often is.

What Are We Measuring Now?

This becomes more complicated when we consider what is actually being evaluated in higher education. If a student can generate a coherent essay, construct an argument, and produce writing that appears polished, then we have to ask what those outputs were measuring in the first place.

It is easy to frame this as a problem of academic integrity, but that framing may be too narrow. The issue is not only that students can outsource work. That possibility did not begin with AI. It is that the system has long relied on products as proxies for learning. Artificial intelligence makes that reliance more visible.

At the same time, students are not the only ones adjusting their practices. Faculty are also incorporating AI into their workflows. They use it to draft feedback, organize ideas, and manage the increasing demands on their time. This creates a situation in which both sides of the educational exchange are interacting with the same underlying systems. Students produce work with AI. Faculty respond with the assistance of AI.


Where is learning situated when both students and faculty are using AI?

Who Benefits and Who Does Not?

The issue of equity further complicates this landscape. Artificial intelligence has the potential to expand access to cognitive support, particularly for students who may not have consistent access to tutoring or other academic resources. At the same time, access to AI tools, as well as the ability to use them effectively, is uneven. Institutional policies vary, levels of AI literacy differ, and some students are positioned to benefit more quickly than others. AI can both extend opportunity and reproduce existing inequalities depending on how it is implemented and who is able to take advantage of it.

Student perspectives on AI often reflect this complexity. Some students use AI to deepen their engagement with material by testing interpretations and refining their thinking. Others use it to complete tasks efficiently and meet external expectations. Both uses exist simultaneously, and neither can be treated as the singular story of AI in education.

AI does not level the playing field. It rearranges it.

Where Does AI Belong?

Efforts to locate AI within a culture of academic integrity emphasize responsibility, transparency, and ethical engagement. These are important principles, but they rely on an assumption that the boundaries of AI use can be clearly understood and consistently applied. In practice, those boundaries are often blurred. As AI becomes more integrated into the learning process, distinguishing between independent work and assisted work becomes increasingly difficult. This shifts the conversation away from detection and toward questions of trust and accountability.

When we step back from individual cases and institutional responses, a broader pattern begins to emerge. Artificial intelligence is not simply adding a new layer to student life. It is redistributing aspects of cognition, altering how students approach tasks, and introducing new forms of interaction into the learning process. At the same time, it is exposing tensions that have long existed within higher education, particularly the reliance on outputs as indicators of understanding.

The problem is not simply how students are using AI. It is what AI reveals about how learning has been defined.

Why the Humanities Push Back

One place where this tension becomes especially visible is in the humanities. Faculty resistance to AI is often framed as reluctance to adopt a new tool, but that framing misses what is actually at stake. Humanities disciplines are built on a particular understanding of how knowledge is produced through close reading, sustained argument, and the development of an individual voice over time. When AI can replicate parts of that process, even imperfectly, it raises questions that extend well beyond academic integrity. It touches the foundations of disciplinary identity and what counts as meaningful intellectual work. What is being contested is not just method, but the definition of knowledge itself.

I explore this dynamic in more detail in a companion video, where I argue that what looks like resistance is often a rational response to a deeper epistemological shift. If you are interested in how faculty are experiencing this moment from the inside, you can watch that discussion here.

What Comes Next

This does not mean that AI should be rejected. There is clear value in its ability to reduce barriers, support idea generation, and make complex material more accessible. The challenge is not whether to use AI, but how to integrate it in ways that support meaningful engagement rather than replace it. This requires attention to how assignments are designed, how learning is assessed, and how students are taught to work with AI systems responsibly.

The more difficult question is not technical but philosophical. If students can rely on external systems to assist with thinking, writing, and problem-solving, then the location of intelligence becomes less stable. It is no longer confined to the individual. Instead, it is distributed across systems, tools, and interactions.

What follows from that shift is still unfolding. Higher education has not yet determined what should remain individual, what can be distributed, or how those boundaries should be drawn. For now, students are navigating that terrain in practice, often ahead of the institutions designed to guide them.

We are not at the end of this shift. We are in the middle of learning how to live with it.