The University of Chicago’s recent AI initiative is a useful signal of where higher education may be heading. AI is no longer treated only as a topic for specialised courses or a tool used by individual students. It is becoming part of institutional life, woven into teaching, research, and the daily routines of academic work. That shift deserves attention because it raises a serious question: how can universities adopt AI in ways that strengthen human judgment rather than weaken it? This goes beyond just being a campus issue. Universities are among the few institutions that still carry a public mission to cultivate reasoning, knowledge, and intellectual independence. If they begin to use AI institution-wide, they are making a statement about what kinds of thinking they value, what learning they want to support, and how much control they are willing to hand over to private systems. Campus AI adoption is therefore more than an administrative decision. It is a choice about the future of human-centred education. The University of Chicago is a revealing example because it has long projected itself as a place of rigorous inquiry and intellectual seriousness. Its AI-empowered research initiative supports a set of interdisciplinary projects across research and education, with leadership framing it as an ambitious attempt to help faculty, staff, and students engage with emerging technologies in a structured way. That already goes beyond the usual cycle of hype, panic, and improvised classroom rules. Instead of pretending that AI does not exist, the university is trying to change how it enters academic life.
UChicago’s own description emphasises that AI will be explored across many disciplines, from the natural sciences and law to public policy, the arts, and education. The initiative funds groups working on how AI can enrich lives through the arts, improve social resilience, support scientific research, and rethink teaching practices. That broad scope makes it clear that the institution sees AI as a topic that intersects with culture, society, and ethics, not just as a technical tool. This is a good starting point. Many institutions either treat AI as a narrow technical resource confined to IT departments, or as a threat to be limited through fragmentary rules. UChicago, by contrast, is treating AI as something that should be examined across disciplines and connected to the university’s intellectual mission. That does not automatically guarantee good outcomes, but it is a more thoughtful posture than simple rejection or uncritical acceptance. The mistake many institutions make is treating AI either as a danger to be removed or as a convenience to be embraced without conditions. A blanket rejection can leave students and faculty unprepared for the tools they will encounter everywhere else, from workplaces to public services. A careless embrace can turn the university into a passive consumer of systems it does not fully understand. Both responses fall short of what higher education can offer. A more responsible approach starts with the question: What is AI for in education? If AI is introduced on campus, it should have a clearly defined role. Is it there to make tasks faster, or to help people think better? Is it meant to reduce friction, or to support students as they work through difficulty in a more informed way? These are not small distinctions. In education, difficulty is often part of the learning process. Writing, revising, searching, comparing, and synthesizing are not simply chores. They are methods through which people develop judgment, creativity, and deeper understanding. The most promising uses of AI in universities are therefore not the ones that only generate polished text. The more interesting cases are those that support inquiry, improve access, and help people work across disciplinary boundaries. AI can summarise large bodies of material, support brainstorming, provide language assistance, and offer new ways of organising research. It can help students who need extra support to guide complex readings, and it can make some aspects of faculty work more efficient when used carefully. In the best cases, AI can lower barriers without lowering standards. This is where a humanistic perspective becomes central. Humanism in the age of AI is not a call to reject technology. It is a reminder that tools should serve human development, not replace it. In a university, this means protecting the role of intellectual effort. There is a difference between a tool that helps someone begin a draft and a system that allows them to avoid learning how to build an argument. There is a difference between an assistant who helps organise sources and a system that encourages dependency. Good institutional design requires that those differences remain visible and that students are taught to recognise them. A human-centered AI strategy in higher education should therefore include some key elements: These are educational questions that require input from teachers, students, and administrators, guided by the institution’s mission. Governance cannot be an afterthought. If a university integrates AI into teaching and research, it needs clear rules, transparent oversight, and faculty involvement. Committees that look at ‘how to think with machines, how to think without them, and how to think about them‘ are a good start only if they have real authority and insight into how systems are used, updated, and evaluated. Faculty should have a voice in setting boundaries, not merely implementing rules designed elsewhere.
There is also the issue of dependence on private providers. When a university builds everyday workflows around a commercial AI system, it becomes reliant on a company’s pricing, product decisions, safety policies, and technical direction. This type of dependence is already familiar from other areas, such as learning management systems and cloud services. AI raises the stakes because it sits close to the act of thinking itself. If a private company becomes deeply embedded in the production of academic work, then the institution should be clear about what it gains and what it risks. Partnerships with AI firms are not inherently problematic. In fact, they may be necessary if institutions want access to advanced tools, training, and support. But partnerships should be guided by public-interest criteria. Universities should negotiate for transparency, educational control, and flexibility. They should think carefully about data use, model changes, and their ability to disengage if systems shift away from educational needs. They should ask not only whether a system is effective now, but whether it can still serve the university’s mission in five or ten years. One encouraging element in UChicago’s initiative is its interdisciplinary character. The projects do not treat AI as only a technical resource. They involve arts, social sciences, law, public policy, and education. This matches the reality that AI has implications well beyond computing. It affects how we organize information, how we make decisions, how we communicate, and how we understand knowledge itself. A university that takes AI seriously should create space for scholars and students from different fields to examine both its opportunities and risks. Legal scholars may look at accountability and rights. Philosophers may explore questions about agency and responsibility. Social scientists may study how AI tools influence participation and inequality. Artists may experiment with new forms of expression. Educators may design ways to integrate AI while preserving the value of slow, reflective reading and writing. UChicago is not alone in treating AI as more than a classroom accessory. Across higher education, a growing number of institutions are beginning to fold AI into teaching, research, and governance as part of a broader rethink of what a university should do in the age of intelligent systems. The details vary, but the underlying impulse is similar: universities are no longer asking whether AI belongs on campus, but are beginning to integrate it. At Imperial College London, AI is being treated as a major institutional priority rather than a specialist side project. Its I-X initiative serves as a shared hub that deliberately brings together diverse fields, embedding AI researchers directly into non-technical disciplines such as medicine, the natural sciences, and business. By connecting frontier research with daily classroom practice, Imperial is treating AI not as a computer science utility, but as a shared intellectual resource for the whole university. The University of Florida offers another instructive example. There, AI has been framed as a central university capability rather than a departmental asset. Backed by major investments in university-wide supercomputing infrastructure and a major cross-disciplinary faculty hiring push, the initiative spans all sixteen of its colleges. Beyond merely offering classes on the technology, the institution has organised itself to support AI-enabled inquiry in fields ranging from the arts to agricultural sciences, demonstrating an ambition to build foundational capacity rather than simply altering the curriculum. A different but equally important model can be seen in Finland’s Generation AI initiative, a multi-university research consortium involving the universities of Helsinki, Oulu, and Eastern Finland. Focused heavily on public-interest research and digital sovereignty, the initiative explicitly demonstrates that AI adoption does not have to mean handing over intellectual autonomy to private, commercial platforms. By developing open-source educational tools that run entirely locally in a user’s browser, without tracking data or exporting it to private tech firms, their emphasis remains firmly on human agency, transparency, and the ability of educational institutions to govern technology rather than simply consume it. Taken together, these examples show that UChicago is part of a wider movement in higher education. Some institutions are moving quickly and broadly, others are proceeding with more caution, but the direction is clear. Universities are starting to treat AI as part of their core academic and institutional life, which means the real challenge now becomes how to do so without losing sight of the public mission of education.
Seen this way, a university-wide AI initiative becomes a test case for human-centered AI. If done badly, it may amount to a procurement story presented as innovation: a subscription to a powerful service, with limited reflection on teaching, learning, or governance. If done well, it can show how institutions can adopt strong tools while keeping their own values and goals in focus. The difference lies in practical choices. Does the university define AI as support for learning, or as a way to automate output? Are students taught to understand how AI works and how to question it, or only how to use it? Do faculty committees only manage access, or do they shape policy and pedagogy? Is the institution transparent about its dependence on vendors, or does it treat AI as just another invisible layer in the background? These are the questions that will decide whether campus AI becomes a step toward more thoughtful education or a slide into comfortable dependency. The message here is not that universities are finally going all in on AI. The deeper story is that institutions devoted to knowledge are being asked to redefine what responsible adoption looks like. The challenge is to use AI in ways that expand access, support research, and enrich learning, without letting the machine become a substitute for the hard, imperfect, deeply human work of thinking. Universities can respond from a place of fear, focusing mainly on cheating and control, or they can respond from a place of confidence and responsibility, focusing on education and governance. They can either outsource judgment to AI systems, or they can teach students and faculty how to work with new tools while keeping judgment their own. If they choose the second path, they will not only be adapting to technological change. They will be showing how human-centred institutions can remain themselves in a time of rapid transformation. That is where courage and humanism meet: in the decision to adopt powerful tools, but to do so under the guidance of human values, educational purpose, and public responsibility. Author: Slobodan KovrlijaA concrete case: University of Chicago
A more careful approach
Humanism and intellectual effort
Governance and dependence
AI as a topic across disciplines
Beyond one campus
A test for human-centered AI
Courage with judgment
