People have long retreated to special places to improve their minds. The library, the monastery, the scientific laboratory—the university itself—has, in all cultures and times, empowered certain kinds of thinking: critical study, critical thinking, scientific experimentation, the pursuit of truth unclouded by political interests. This program supports the residential college model, a focused environment that promises to transform young people through rich learning experiences that occur in all areas that support exploration, inquiry and focus. In 2026, we are forced to ask, what happens to the different cognitive experiences offered by residential college in the age of reproductive AI?
I share this from the front lines: College instructors are busy. After twelve years as a professor of Spanish, I transitioned in the fall of 2025 to work closely with faculty at all levels as an educational engineer at a private liberal arts university, where I provide training to support teaching expertise. During the fall semester, teachers explained to me the dramatic rise in unauthorized use of AI. By January 2026, many officials had concluded that the written assignments that had been the mainstay of their teaching—providing a meaningful way to assess students’ reading comprehension and develop their analytical skills—were no longer worth giving. Students were submitting AI-generated prose. The math industry began to find it impossible to give students points for completing homework sets that were apparently released from a chatbot.
Much has been made of the return of the blue book test to the classroom. In relation to the amazing face of recent practice, faculty have students complete important assignments by hand in class. This solution, however, creates a new problem: lack of time. A Spanish professor asked for my help in redesigning a short fiction class for the Spanish language. In recent semesters, his advanced Spanish students had begun to struggle more with understanding short stories. They would complete assignments at home, apparently relying too much on AI, but they wouldn’t be able to do those skills in the classroom. “They can’t learn Spanish anymore,” he worried.
I suggested that she use class time to hold workshops, coaching students through the challenging process of developing comprehension in learning a new language. “I only have 75 minutes, twice a week,” he replied. “I don’t have time for that.” You are right. In our current system, faculty have students one-third of the time they should spend on classroom learning activities. If we can no longer rely on students to engage meaningfully with learning outside of the classroom, up to two-thirds of their learning time is lost. The correct fix is to increase class time, but for reasons beyond the scope of this article, this change isn’t coming anytime soon. On a practical level, many institutions do not have the space to accommodate long classes.
While there is no single line of defense that will prevent the intrusion of AI, my conversations with faculty have led me to believe that an important part of the solution is creating a new kind of space on campuses—one without AI.
Imagine a place where students can pursue a serious learning task, without the temptation of AI shortcuts. To make this vision a reality, I propose the creation of human intelligence labs, spaces designed to foster the intellectual interactions that often distinguish advanced studies. Whether located centrally on campus or housed within academic departments, human intelligence labs can be spaces designed by staff to allow the use of technology that supports student learning, while limiting access to those who undermine it. Most importantly, these spaces will enable faculty to block access to AI, forcing students to practice reading, writing and problem-solving skills fundamental to their fields.
Revitalizing the physical campus—reimagining where and how learning happens in all areas of campus—is a major undertaking that will take time and resources. If residential colleges are to continue to deliver on the promise of providing a transformative learning environment, however, we must think creatively beyond current programs and structures.
As a professor of Spanish, I position the language lab as a jumping off point for what we create. If you studied a modern language in college, you probably went to the computer lab to complete assigned listening and speaking exercises. Language labs are disappearing as students gain direct access to multimedia resources through their devices. However, in its time, the language lab served as an important classroom where students engaged in effective struggle with the target language. College campuses currently boast a variety of related spaces, such as writing and tutoring centers, that support student learning outside of the classroom. Our challenge is to change those places and build on their success to create new places that respond to the main challenge of the day—the penetration of AI.
Consider the possibilities. Students in fields ranging from history to sociology can visit human intelligence labs to write research papers using only the technology their professors deem appropriate for the task. They would express and clarify ideas, form arguments, and make changes—mind-intensive tasks that students might be tempted to offload to AI. Students visited human intelligence labs to complete many of the normal higher education tasks that you and I have completed in our dorm rooms, the library or the campus coffee shop; Our current technological environment, however, requires us to build a strong infrastructure that protects these structures of the learning process.
Sadly, these labs will allow faculty to continue to assign research and analytical papers, work products that many currently avoid due to their vulnerability to unauthorized use of AI. Human intelligence labs can be the on-campus brain gyms that Cal Newport recently described.
Human intelligence labs can also encourage deep engagement with learning, an effort that has hurt creativity as both the video-friendly media landscape and the availability of AI abstractions threaten to make steady, deep learning obsolete. Faculty can try different ways to strengthen students’ learning muscles. A low-tech approach might involve assigning students to write hours in a lab similar to a regular classroom, equipped with comfortable chairs and tables, while leaving devices in hallway closets. A high-tech approach might include a computer lab where students study assigned texts on computers equipped with public annotation software such as Perusall or Hypothes.is, but block access to generative AI. (Yes, the faculty discovered that students were using AI to complete the annotations required to study the course.)
Math students can benefit from this same low-tech environment, completing problem sets without struggling with the lure of omniscient chatbots. If students are assigned specific attendance times, as they are with chemistry labs, they can engage in peer learning or lab sessions can be attended by teaching assistants or instructors.
For human intelligence labs to function effectively, they will need to hire trained staff to monitor compliance with established protocols, such as leaving equipment in cabinets. This can be the most controversial part of the model, rekindling the debates about monitoring our students (using proctoring software) that arose during the throwing of emergency remote instructions in times of violence. The Faculty does not want to build relationships with students on the basis of suspicion and distrust.
There is evidence, however, that college students want us to protect them from invasive technologies, including AI. If we don’t create these spaces, they create themselves. For example, at the beginning of this academic year, a group of students from St. John’s College in Santa Fe organized a six-day “technology fast”, locking their phones and other things in their classmates’ suitcases. Students expressed a desire to feel more connected to each other and less addicted to gaming apps and immersive scrolling effects. Human intelligence labs can answer students’ hunger for freedom from the grip of technology.
Although supervised, human intelligence labs will be very different from testing centers. Whether run by college campuses or by third-party companies like Prometric, testing centers focus on summative tests, which measure the extent to which test-takers possess knowledge and skills. Their goal is to create a secure environment for honest testing; surveillance is at the heart of what they do. The goal of human intelligence labs could be very different: to serve as a warm, welcoming environment that supports students in using human intelligence.
Of course, AI should be embraced in the right places on college campuses. I met recently with a faculty member from the Institute of Entrepreneurship who allows their students to access AI without limitations in many jobs because AI, in his words, is an entrepreneur’s best friend. Employers in the enterprise space are looking to hire college graduates who are advanced users of advanced AI tools. If anything, this professor was concerned that his students’ use of AI was too basic for the current needs of the job market. He gives his students weekly “AI challenges” to challenge them to use the tools in new ways. This professor may use human intelligence labs sparingly. For professionals in many other fields, a space without AI can be a godsend, allowing them to focus on teaching their discipline rather than spending all their time trying to outdo AI.
Institutions using human intelligence labs will undoubtedly face many challenges. Creating flexible, AI-free environments to engage in complex environments, technology and accessibility planning. Institutions will have to modify existing facilities or build new ones. Faculty will need to discuss and agree on the terms of those positions, including supervision standards. It will be necessary to hire staff. Departments will need to obtain information related to this change. Most students access textbooks as ebooks, so we’ll need to find workarounds (one idea: Stock the lab that math students often access with hard copies of the textbook). All of this will take time, innovation and money. Indeed, higher education investment in resources will need to match—in design and intent, if not in raw numbers—Big Tech’s almost incomprehensible investment in AI, which is projected to total $700 billion by 2026 alone.
The arrival of the latest models of major languages on our campuses has felt, for many, like a huge kudzu vine, hell bent on covering all areas, wrapping itself around our most important teaching tools (the research paper, the creative project, the humble reading assignment) and strangling them until they are lifeless. Faculty cannot deal with this criminal simply by teaching differently. We cannot expect students to refrain from using these attractive tools. A thoughtful, multifaceted institutional response is needed to preserve the value of the residential college, and to protect the teaching tools and techniques that serve our students and alumni. The creation of human intelligence labs would be a logical step in that situation.
