Across higher education, artificial intelligence is too often being governed as though it were primarily an academic integrity issue. It is clearly not just that.
AI is already reshaping how universities teach, advise, recruit, admit, communicate, assess risk, and make decisions. Yet many institutions continue to approach it through fragmented policies, uneven faculty guidance, and conversations narrowly focused on misuse in student work.
That is a strategic gap our industry will soon regret. AI is rapidly moving beyond the classroom and into the core of institutional operations.
This important shift demands attention not only from faculty, but from within senior leadership and governing boards. Universities that fail to establish a coherent, enterprise-wide AI strategy, supported by appropriate technical architecture, risk more than policy inconsistency.
They risk erosion of brand relevance, weaker market positioning, declining institutional trust, and, over time, measurable impacts on competitiveness, enrollment, and productivity. As higher education enters a period of accelerating disruption, these risks may prove existential for some.
What it means to think
At the same time, students are not waiting for institutional clarity. They are already adapting, often quietly, to a world in which AI is becoming an ambient layer in how we think, learn, live and work.
Many institutions still frame their response in terms of “AI literacy.” Students, however, are rapidly developing something more consequential: practical fluency in seamlessly working with AI as a cognitive partner in this emerging age of co-intelligence.
This distinction is profound. Literacy implies awareness and basic competence. Fluency requires judgment, adaptability, and the ability to engage dynamically with AI as part of one’s reasoning and problem-solving processes.
The institutional challenge is no longer whether students will use AI. It is whether universities can define what it means to think, learn, and demonstrate relevant knowledge in a world where AI is embedded in the complex process of thinking itself.
Much of higher education’s current response risks missing the nuance of this inflection point. AI detection technologies remain imperfect, yet are still relied upon in some contexts as though they were definitive.
Meanwhile, internal policy debates often reflect a search for institutional comfort rather than an acceptance of reality, ranging from cautious experimentation to outright prohibition, as if AI is an online destination we can ban rather than a pervasive technological shift that is reshaping cognition itself.
The result is fragmentation, with students and faculty navigating inconsistent expectations in the absence of a clear institutional north star.
AI fluent administrators needed
This is, at its core, a governance problem. Many institutions are still attempting to manage AI through a single policy lens, aligned more to institutional preference than to the scale of the transformation underway.
That approach is unlikely to hold. Effective governance requires a clearer distinction not only between policy and practice, but between academic and administrative uses of AI, each of which introduces different strategic, operational, and ethical considerations.
It also requires stronger AI fluency within senior leadership and governing boards. Without informed internal perspectives, institutions risk defaulting either to inertia or to vendor-driven narratives that may not align with their long-term interests.
In academic settings, AI governance must address authorship, disclosure, collaboration, and assessment. In administrative settings, it raises different concerns: data governance, privacy, procurement, bias, institutional decision-making, and risk accountability.
These domains overlap, but they are not interchangeable. Conflating them produces reflexive over-reaction, weak policy, uneven implementation, and institutional confusion.
A single, generalized “AI policy” will not provide the clarity universities now require, particularly when the underlying technologies are evolving so rapidly that static policies risk becoming outdated almost as soon as they are published.
Governance must therefore become more adaptive, and more responsive, even when that pace feels uncomfortable. The goal is not comfort. It is leadership.
Preparing for an AI ecosystem
A brief example may help. I recently created a first-year seminar in which students did not simply use AI, but reflected on when and how its use strengthened or weakened their thinking. Many expressed anxieties about privacy, over-reliance, and the risk of being falsely accused of misconduct in environments where policies remain unclear.
At the same time, they recognized that nurturing fluent AI capability was already essential to their future employability. Through structured exploration and sharing standards for responsible use, that tension gave way to more deliberate engagement.
Students became better able to distinguish between original thought and meaningful collaboration, using AI as an aid to thinking rather than a substitute for it.
The lesson is not that AI should be embraced uncritically, nor that it can be managed through simple restriction. In the absence of institutional clarity, students will either avoid AI out of fear or use it in ways that diminish their reasoning over time.
Neither outcome serves the mission of higher education. And our goal as officers and governors should be to protect our mission.
Nor is the answer to select a single AI platform and declare the institution “AI-enabled.” That is not strategy. The leadership landscape among models shifts rapidly, and no single platform will meet every institutional need.
Universities should instead be preparing for an AI ecosystem, one that includes multiple models, embedded applications, and, increasingly autonomous agents, each serving different purposes within a coherent institutional framework.
3 big AI implications
Artificial intelligence is not another tool to be added to our office technology stacks. It is an embedded strategic capability that will reshape how work is performed across the university. For institutional leaders, the implications are immediate.
First, universities need separate but coordinated governance structures for academic and administrative AI use cases. Faculty governance alone cannot address enterprise risk, just as administrative policy alone cannot define academic integrity. Coherence is essential.
Second, institutions should define AI fluency, not merely AI literacy, as part of their academic mission and market relevance. This requires preparing students not only to use AI tools, but to exercise judgment, ethical discernment, and meaningful human oversight in working with them.
Third, AI must be treated as an enterprise imperative, not a temporary disruption. Decisions about procurement, data use, policy, training, and implementation should be aligned across the institution rather than allowed to emerge unevenly across departments.
