Every Nigerian undergraduate now has access to AI tutoring. Almost none use it well. The default prompt — type a question, copy the answer, move on — is the same instinct that made copying from a friend in secondary school feel productive. It gets you through the assignment. It does not get information into your head.
The students who actually learn from AI tutors talk to them differently. They use the AI as a Socratic interlocutor, not as a homework oracle. The shift is small in keystrokes but enormous in outcome.
The four-question framework
- Explain this concept to me as if I have never seen it before.
- Give me one worked example.
- Now give me a similar problem to attempt — do not show me the answer until I have tried.
- Here is my attempt. Where did I go wrong, and what should I do differently?
This four-step sequence forces engagement at every stage. You read the explanation. You see one example. You attempt a new one. You get diagnostic feedback. By the end, you have done the cognitive work of learning, not just consumed information.
What to ask versus what not to ask
Good AI tutor prompts ask for explanations, examples and worked problems. Bad ones outsource thinking. Compare these two prompts on the same topic:
Bad: "Solve question 3a for me." · Good: "Walk me through how to approach question 3a, show me the first step only, then let me attempt the rest."
The first prompt gives you an answer. The second teaches you the method. On exam day, only the second one helps — and the exam paper will rarely be question 3a anyway.
Asking for the underlying logic
When you do not understand why a formula works, ask. AI tutors are excellent at explaining the intuition behind a procedure — what each step is really doing, why the conventions exist, where the formula breaks down. Knowing the logic is what lets you adapt when the exam question phrases things differently from the textbook.
The cross-examination test
A useful integrity check on whether you are learning or just collecting answers: can you re-explain the topic out loud to a coursemate without looking at the AI conversation? If yes, you have learned it. If no, you have only seen it. The distinction matters because exams test recall and application, not exposure.
Used the right way, AI tutoring is the most personalised and patient teaching most Nigerian undergraduates have ever had access to. Used the wrong way, it is faster cheating that still results in failed papers. The students who graduate well in the next five years will overwhelmingly be the ones who learned to use these tools as tutors — not as homework machines.
Share this article
