A group of students sitting round a table. They have coffee, laptops, and books and they seem to be chatting.

Generative AI products like chatbots, writing assistants, and AI‑enhanced search tools are freely available now and it can be very tempting to use them in a whole range of situations. The key thing to remember when you are considering using them is that you have enrolled on a course to learn. Whatever subject you study, your teachers design the courses to help you learn, develop your analytical and communication skills. When they set an assignment, they want to know what you can do, and to hear your voice. If you are having difficulty with this, your first stop should always be your teachers so that they can interpret your questions in the specific context of your course.

Read this short article to set this into context: Greene, J. A. (2025). Learning With GenAI: Shortcuts Versus Cutting Yourself Short. Psychology Today. There is a concept known as "desirable difficulty" in learning. Desirable difficulties are activities which are challenging, like testing your memory and answering quiz questions without reference to notes or resources, which have been found to improve your long-term retention of information and concepts. If you just reach for an AI answer instead of stretching yourself by engaging with these desirable difficulties, you won't achieve the kind of long term knowledge and understanding which is expected of a higher education graduate.

That doesn't mean that GenAI products have no place in your study routines. Instead of replacing your thinking, these tools work best when you use them to support your learning: helping you understand difficult ideas, plan your work, ask better questions, and practise new skills. Used with understanding and intention, these products may help you learn more efficiently and confidently. Used uncritically, they can give you the illusion of learning, confuse you, oversimplify things, or even give wrong information.

GenAI creates text, images, code, or explanations based on patterns in large datasets. It is designed to sound convincing, not to check facts. This means that it can:

be great for inspiration and practice.
produce made‑up information which you need to check.

You have to remain the decision‑maker. Your judgment, your skills, and your understanding still matter most, especially because you are responsible for all the academic work you submit.

GenAI can support you at different stages of learning:

Before you start: Use it to get an overview of a topic or generate questions you should think about.
While you work: Ask for explanations, examples, or alternative ways of expressing your ideas.
After you draft: Get suggestions to improve clarity, structure, or flow—and learn from the changes.
When revising: Turn your notes into summaries, flashcards, or practice questions.

The goal is not to let AI take over your tasks, but to see if it can extend what you can do. Take a moment to think about your own study habits:

Which tasks usually take you the longest?
Where do you often get stuck: understanding, planning, writing, or revising?

Could GenAI help you with any of these without interfering with you actually learning? Make a private list of ways you might use GenAI products in your studying. Now have a look at this list, which covers all kinds of different ideas for using AI in Education AI in Education: 39 Examples. Do you want to add anything to your own list? Are any of these examples new to you, and interesting? Do any of them seem like a bad idea?

Plenty of people have opinions about the negative impact of GenAI in the classroom, and the truth is that we don't have enough research available to know. If you are interested in learning more about this, check out this YouTube video by Cole Hastings about the challenges of GenAI in education. You need to think about yourself and your own learning. Do you feel like you are offloading your development to a product, or using a new tool in a smart way to make you more efficient? Are you using GenAI to deepen your understanding, or to give you quick answers? Be honest with yourself, and remember that you know best when you are learning and making progress.

Keep your list of potential applications and come back to it in the next section when we think about ethical approaches.

Further reading

As you will find out in a later page, research into how GenAI impacts learning is very early and there is only reliable data from very small scale controlled trials. But these articles give some sense of how we might respond as humans to always getting an easy answer (even if it is wrong!).

Barcaui, A. (2025). ChatGPT as a cognitive crutch: Evidence from a randomized controlled trial on knowledge retention. Social Sciences & Humanities Open, 12, 102287. https://doi.org/https://doi.org/10.1016/j.ssaho.2025.102287

Bastani, H., Bastani, O., Sungu, A., Ge, H., Kabakcı, Ö., & Mariman, R. (2024). Generative AI Can Harm Learning. SSRN. https://papers.ssrn.com/sol3/papers.cfm?abstract_id=4895486&s=03

Poulidis, S., Bastani, H., & Bastani, O. (2025). Self-Regulated AI Use Hinders Long-Term Learning. Available at SSRN 5604932. http://dx.doi.org/10.2139/ssrn.5604932