In Spring 2026, a randomly selected half of the student body received a survey comprising three NSSE Shorts modules, one of which asked about their perceptions and use of generative AI tools. Overall, 529 students (34% of those invited) responded to the survey. This report summarizes the comments shared by 374 of these respondents to the questions: “What are your thoughts about using (or not using) generative AI tools for learning? What factors have influenced your decision?” Commenters were slightly less likely to use AI compared to survey respondents overall; 10% reported using it “very often” or “often” (compared to 13% of all survey respondents), 56% used it “sometimes” or “rarely” (compared to 59% overall), and 34% “never” used it (compared to 28% overall).
Aside from light editing to spelling and grammar, quotes shared here are copied verbatim. Many comments addressed more than one of the themes discussed in this report. Sometimes, just excerpts are shown to highlight a specific theme.
Accepted Uses of AI
More than half of students’ comments (53%) described particular circumstances where they felt generative AI use was acceptable or helpful. Many students qualified their support for AI use with phrases like, “when it is used correctly” or “ethically” or “responsibly,” but there was not necessarily a common consensus around what this meant. In general, students seemed supportive of using generative AI for tasks such as summarizing material (particularly dense or jargon-heavy texts), explaining difficult concepts, creating study guides or practice exam questions, brainstorming essay or research topics, or providing feedback on a completed assignment. Still, not all students found each of these uses acceptable. Finally, a few students discussed uses outside of the classroom, such as job searching or personal hobbies.
- “AI is okay to use on your own for studying to reinforce concepts, not as a replacement for your brain. I already have my own notes on concepts which I made by going to class, and AI helps me reinforce those, not come up with them.”
- “I don’t think AI should be used on tests, BUT if your professor is assigning homework without teaching the material on the homework, it is okay to use AI. Or if you’ve tried a problem and cannot get it even when using other resources, then it’s okay to have AI explain it to you. But never just copy down an answer, read it and take notes and understand it. If you do that then you have done what any professor can expect.”
- “I think AI can be used well in order to help summarize readings that would otherwise take too long or there isn’t enough time for, when it comes to using it for assignments and testing it is unethical but it isn’t the worst when it comes to using it for saving time on a 70 page reading or helping you study for an upcoming test because being able to generate practice questions from a lecture or a reading can help you identify areas of struggle.”
- “I only use AI to help find errors in my work or give me suggestions for improvement. I feel like this is a responsible and honest use of AI, given that the work is still mine and AI is being used as a peer editor. However, I don’t think most students use AI in this way.”
- “I think AI does a good job of answering questions with in-depth reflection that if I need certain steps broken down more, I am better able to understand as a student. I think having this deeper understanding allows me to apply certain knowledge in future problems … I do think it is okay to have AI check your answers to homework problems to make sure your thinking is on track–especially when a lot of professors don’t allow revision on homework problems. For me, homework is the time to make errors which you can learn from for the test.”
- “For me the hardest part of any writing assignment is brainstorming or narrowing down a topic and I have sometimes used AI to help brainstorm which has been very helpful and made the process significantly shorter … I have found AI also helpful sometimes in explaining some complex topics that would otherwise be difficult to grasp.”
- “I think that it requires a certain type of learner to use it effectively. For instance, as a curious person I use AI to help me understand where things are fuzzy, and I rarely use it to take shortcuts. I think it can be a great, accessible tool if used with the right intentions.”
- “I think that there is a time and place for AI use … as a busy college student I find it helpful for summarizing a job description and related tasks. I try to minimize my use in academic settings. However, it can be really good at explaining difficult chemistry concepts in a straightforward way. I have used it to explain how a certain answer was reached.”
- “I think there’s honestly a bit too much negativity around AI. It really can be used in some really helpful ways. It is a very powerful tool that when used in the wrong way can definitely hurt your studies, but this type of tool has existed since we were kids — the internet. These chatbots are just a new way of interacting with this information. Students should certainly use self control when using AI, but it can be incredibly helpful. Personally, I’ve used it to give me feedback on my essays. To be clear, I never ask it to revise or write any of it for me, I just ask it for feedback and to check for grammatical errors. It’s been incredibly helpful.”
- “Using [generative AI] to make workload more efficient is the best thing that ever happened to college students in my opinion. In a world where everyone is constantly overstimulated and bombarded with information while having to participate in life’s everyday responsibilities, having ChatGPT work out your calendar for you or taking care of minute details could be useful, so long as the student understands and acknowledges the ethical, environmental, and cognitive implications generative AI has on the world.”
Concerns About AI
As is already evident from the quotes shared above, nearly all students (even more frequent users) gave qualified and nuanced responses about when or whether generative AI should be used; only four comments offered no concerns about its use. On the other hand, 87 students (nearly a quarter of all open-ended comments) expressed strongly negative views about generative AI, viewing its use as unethical or antithetical to an educational setting. Many students, including those who do support using AI in certain circumstances, expressed ethical concerns; these are discussed further in a later section.
- “I do not believe people should be using generative AI tools for learning. Although I understand it might have some efficient uses, is using the water worth it? Is the burning world worth it? In my opinion, no. Thinking on your own is now brave and radical.”
- “Generative AI is never useful for learning. There are plenty of resources that do not actively harm the environment and communities and that do not literally make you less smart. These ‘tools’ are an insult and danger to human creativity, learning, and connection. Factors include how I have gone my whole life without it and am fine with the resources provided to me, and through my learning believe that any benefit it supposedly creates for access is heavily heavily outweighed by its ethical failures. Generative AI has no place in my life or my education.”
- “Generative AI should never be used in learning, with the exception of when learning about ethics of AI. If there is an assignment where the professor is considering including an option that uses generative AI and one that does not, the option that does not should always be the only option.”
- “I believe that AI should not be used for learning especially in college when students are learning to think for themselves. AI takes away from independent thought and creativity. I believe this because I have seen the effects of using AI on my peers and how they can not always do simple tasks without using AI.”
- “I believe it is both incompetent and immoral to incorporate generative AI for one’s own learning experience. The path most trodden is often the least effective path, and that is incredibly apparent with such tools as “Grok” or “Sora”. Personally, it is necessary for one to cultivate their own path, to strive forward with their own momentum and passion.”
- “I think it is extremely unethical and embarrassing to use generative AI for anything. I hate generative AI and think St Olaf should be stricter about using it. Most people on campus in my experience also hate generative AI. The factors that influence my decision are knowledge of the harm to the environment, knowledge of the unethical data scraping these companies do that steal work from people, knowledge of how these companies react to people literally dying from using their product, knowledge of how these companies treat the data and privacy of their consumers, and just being a person that hates the way all of this has been shoved down our throats”
These comments were slightly more likely to come from juniors and seniors (39% came from these classes, compared to 33% of all comments), suggesting that younger students may be more open to using AI as it becomes more ubiquitous throughout education. Several additional students said that they “never” use AI, but framed this more as a personal decision rather than a censure against all AI use. And almost two dozen of those strongly opposed to using AI still reported using it “rarely” or “sometimes” during the current school year; perhaps they had formed their strong opinions later, or were required to use AI for coursework.
Impacts on Learning and Academic Integrity
In addition to those strongly against the use of generative AI in all cases, many students shared objections to particular uses in an academic setting. These generally converged on two themes: concerns about learning loss (48% of comments) and/or objections related to academic dishonesty (32%), such as using AI to fully generate essays or complete exams. Several students expressed a desire to wrestle with difficult material on their own and saw this struggle as essential to learning. Some shared fears that they would come to rely on AI as a “crutch” if they used it too frequently; as one student amusingly put it, “you would not take a forklift to the gym.” Others took a more practical stance, arguing that using AI to complete assignments was not a good return on investment in their education. Finally, some students felt that if their peers used AI to complete assignments while they did not, it created an unfair disparity or took away from the value of the learning experience for all.
- “I personally believe using AI in an academic setting will always hinder learning if it is used as a shortcut. I try very hard to resist the impulse to turn to AI when I’m confused and instead force myself to handle the discomfort of not being able to find an answer, even if that means I’m unable to get a 100% on an assignment. I do this because I believe it is the only way to keep my critical thinking muscle from atrophying in the way I have seen occur for many others because of overreliance on generative AI whenever they experience any discomfort academically.”
- “Learning is not strictly a noun, it is a verb – it is a process. Therefore using GenAI, in the vast majority of cases, bypasses any meaningful self-application. For all intents and purposes, as students use GenAI, it does not support learning, it supports task completion.”
- “I think it can be a very powerful tool, and I don’t want to get left behind for not knowing how to use it. However, I’m scared of its potential to weaken critical thinking and problem solving skills. I don’t want myself or the larger community to become dependent on it because then I think so many other problems will pop up. I’m honestly not sure the best path forward.”
- “I think using generative AI to test knowledge and as a tool for learning is acceptable, but I have often seen it turning into a crutch for people. Seeing people use generative AI to complete simple assignments has made me realize if one is not careful, it can impede your ability to learn and create rather than support it.”
- “I think relying on a technology to make things easier or simpler defeats the purpose of learning; it detracts from learning how to synthesize, reframe, etc. on your own as a human. Factors that influenced me include using AI for such purposes and noting that I started to struggle to learn (or fail) on my own, and that I didn’t retain anything that I thought I “learned”. Additionally, I don’t like taking the human (including the mess, the confusion, the mistakes) out of learning.”
- “I do not believe generative AI should be allowed to complete any coursework that receives a grade. I do think there is some benefit to being able to ask questions about a scholarly text to further your understanding of it, but never to complete work for your own benefit.”
- “I think it is ethical to use AI as a tool to make learning more efficient. For example, summarizing a very long reading to get the main ideas, or explain step by step thoroughly how to solve a difficult calculus problem. However, it is very unethical to use it as a method of cheating, i.e. using it to get answers for a test or assignment.”
- “AI is removing the innate human capability of critical thinking. I feel like I am watching my peers become dumber. We pay and go to college to learn and grow, why are you having AI do it (and do it badly) for you”
- “I stay clear of AI for the most part because I am here in order to strengthen my own thinking, writing, and reading ability. If I wanted a machine to do those for me I wouldn’t pay so many thousands of dollars a semester.”
- “Even if students do not use AI for assignments/cheating etc. I have frequently seen students using it to summarize assigned readings (in lieu of reading them for themselves) … This is concerning to me because readings assigned in discussion based classes promote the open sharing of ideas. It is not only damaging to the student who uses AI, but also the students they converse with because it is not a genuine, personal reflection. If you are not completing readings it is harmful to your learning AND those around you … I have the incredible opportunity to attend college with students from all over the world, with such diverse perspectives, religions, cultures, and ideas; I feel that it is a disservice to all students to encourage the use of AI over the knowledge that each student has in our community.”
- “I came to college to take classes with other humans who are invested in refining their critical thinking skills. I came to a college that has largely discussion based classes, because I want to engage with other humans on their point of view. If my classmates are largely using AI to come up with their view points, they miss out on the nuance only a human can express. They are not walking away with thewhole picture, they are walking away with the spark notes.”
- “It feels very impersonal, inauthentic, and non educational to be receiving artificial ways to function. When messaging my professors I want to know I am connecting with them personally, not receiving a message written by a computer. I feel the same when putting much effort into a group project and a partner uses AI to “save them time”. They’re not learning or practicing the content when something else generates it for them. They’re not learning how to properly work as a team and ask real people and professors questions. It feels immoral and inauthentic to use ai. I connect with my peers and professors better when I and others do not use it. I feel this is what a liberal arts college is supposed to teach us. There is more than just our own views, major, and efficiency.”
Ethical Concerns
Nearly a third of students (32%) shared various ethical concerns about using generative AI; these included its impact on the environment (shared by 28% of respondents), viewing AI use as plagiarism (8%), and believing that AI companies and tools exploit vulnerable populations or produce biased results (7%).
- “AI is incredibly destructive for our brains and the environment. There are hundreds of studies showing that we are destroying our planet with the amount of water and energy AI data centers require, and studies showing that AI hurts brains by replacing critical thinking, creativity, and writing skills. AI should never be used.”
- “I personally do not use it because I feel no need to use generative AI to enhance my learning aside from what I am already doing to succeed in my classes. I also recognize that AI tools can consume valuable natural resources, such as water, so it feels extremely unnecessary for me to contribute to that waste with AI queries.”
- “I believe that it is possible for generative AI to be useful in educational settings, BUT I think the environmental and ethical issues make it so the negative far outweighs the positive. I think it is unethical for the institution or its professors to advocate for the use of generative AI.”
- “Ethically, I am strongly against the usage of AI. The environmental costs of data learning centers are felt the hardest by low-income communities, it discourages critical thinking and problem-solving skills, and I’d rather make bad art than have a computer make my art for me. There’s also the fact that it takes from a body of resources (especially artists and writers) who did not consent for their work to be used, and therefore I view it as stealing to use generated work built from that content.”
- “Generative AI is essentially a plagiarism generator. It constructs its responses based on others’ work that it may or may not reasonably cite … Additionally, GenAI strains and, as its use becomes more common, bypasses ethical considerations of and alienates students from the use of GenAI. Students do not think about the process by which data was scraped, algorithms were developed, and products catered to their (non-)specialized use. I believe this results in a poverty of both civic engagement and responsibility as academics.”
- “Large tech companies control the information that AI models gather, allowing bias and misinformation to blossom in this unregulated online space. The environmental degradation associated with water cooling needs and mineral extraction is also almost never mentioned, which I feel is a very compelling reason to avoid AI almost entirely.”
- “I think the use of generative AI is a form of plagiarism, at least for my classes and assignments. Even in situations when it can be ethically justified, I don’t feel confident about its accuracy, having heard stories about faulty information and responses, and would prefer to complete assignments myself. If I do well or poorly on an assignment, I want to know that my grade was a result of my own merits or mistakes.”
- “I think it is extremely unnecessary in most areas of study, especially in writing. If any professors are encouraging the use of AI or allowing it, they should be required to educate students on the environmental implications and the biased data it is programmed on.”
Performance Issues
Several students (17%) expressed frustrations with inaccuracies or hallucinations from generative AI tools, or simply did not find these tools useful. Some students believed that they should be used as a “last resort” after trying other academic support resources.
- “AI tools can be helpful for self-grading assignments before turning them in and providing good feedback on essays, but I find that when AI attempts to rewrite my work, it doesn’t sound like me and often makes stupid mistakes.”
- “Given the risks of inaccurate and/or hallucinated results, you kind of need to already be competent in a subject to use AI in it so that you can check its work. However, I do suppose there could be some beneficial use of it to help your learning when used in the right way (to help you learn to understand course content, not as a crutch for giving you the answer), though the reliability issues would make me hesitant to trust anything gleaned from it in such a situation”
- “Watching AI destroy the truth and aid in the spread of misinformation makes me believe that it is not to be trusted for any information, relinquishing it to roles in proof-reading, grammatical checks, and other tasks of the sort.”
- “Well it just doesn’t do anything meaningful. It’s not very good at writing; I rarely ever need to formally “summarize” things, and I would do it better without AI; I don’t need its ideas when I can think of better ones myself; the list goes on. AI doesn’t aid learning, it just makes it more “efficient” – or, rather, it makes completing classwork tasks more efficient. These are not the same thing. Given the myriad ethical nightmares surrounding the technology, I see no good reason to use it for learning.”
- “As long as it’s being used as a tool (thesaurus, tutor, etc.), and NOT to cheat, I think it’s okay. I think it would be better as a last resort, though. Explore the resources we have here at St Olaf — Library, Disco, SI, Office Hours, etc.”
- “There are many many websites to look through if you need something, and AI is biased and also wrong a lot of the time, so it does not even help students because I know many students do not double check what their information is.”
Lack of Regulation
Finally, about 9% of comments discussed hesitations around using generative AI given a lack of regulations or guidelines. In this context, some students referenced similar environmental or other concerns as above. Others recognized the inevitability of this new technology, but were unsure how to use it responsibly. Additionally, some students wanted to see more guidance specifically from professors or the college about when it is appropriate to use AI.
- “I believe AI is going to lead to a decline in human intelligence and curiosity and I also do not think it should be used until there are federal regulations as well as a plan to change the resource consumption that currently impacts the environment and poorer communities.”
- “Because of its ethical and environmental connotations, generative AI, though it is an important tool to understand, is difficult to justify in an academic setting. Its lack of regulation, excessive pollution, and the fact that it’s trained with stolen data and artwork make generative AI a dangerous tool that is easily and often abused. It is not sustainable in its current form, and it needs significant reform before it becomes something I seek out.”
- “I think if we don’t use AI in school and then get into the real world where I know many jobs have embraced AI, we will be ill-prepared. When talking with adults who use AI in workforce, I worry St Olaf professors are being too strict about AI usage to the point where it’s this St Olaf exceptionalism that won’t serve us well in real world, and this is coming from someone who hasn’t used AI for school nor likes AI, but I am also a realist and recognize that the St. Olaf policies need work.”
- “AI has far too little regulations at this time, so if the college wants to encourage its use, AI policy and regulation must be made with the input of everyone. Transparency and the freedom to stand by your own values must be considered in this, as well as the different ways the educational departments might view/use AI.”
- “I don’t think it is practically possible or even reasonable to ban AI from institutions of higher education, but I do think we need some kind of regulatory and/or moral system that disincentivizes its use as a pathway to offload the very learning we are here to do. There is a very real conversation to be had about teaching people how to use AI, as it is already being implemented in nearly every discipline and industry. But I believe we need to separate AI as a tool to complete work or solve problems, and AI as a way to avoid learning. Of course, every individual is ultimately responsible for their own education, but there is still plenty that can and should be done to help.”