How to study with AI using your own course material

By the Sendy team, students who built Sendy | Updated October 14, 2026

To study with AI, give it your own course material (slides, readings, lecture recordings and the syllabus) instead of asking it general questions. Then use it to test you, not just to summarize: practice questions and flashcards built from your files make you recall the content, and recall is what improves exam results.

Why your own material matters

A general chatbot answers from what it picked up on the open web. Your professor teaches a particular version of the subject: their notation, their examples, the topics they chose to stress, the way they word exam questions. When the AI works from your files, its notes and questions match the course you will be graded on.

The syllabus matters more than most students expect. It lists what each week covers, how the course is graded, and when the tests are. An AI that has read it can tell you what is on the midterm and write questions in the same format.

An economics class in Sendy: two units with their sources, notes and podcast, and a How the weeks connect summary showing Unit 3 builds on Unit 2 — With the whole class uploaded, each week is linked to the ones before it. From the demo account.

What to give the AI

The syllabus. Topics by week, the grading scheme, and the exam format.
Lecture slides. The outline of what was taught. Export Google Slides or Keynote to PowerPoint or PDF.
Readings. The textbook chapters and articles assigned for each week. They hold the explanation the slides leave out.
Lecture recordings. What the lecturer said out loud, which is often the part that ends up on the exam.
Past tests and problem sets, if your course shares them. They show how questions are asked and marked.

Keep each week's files together. Notes on week 6 are better when the AI also knows what weeks 1 to 5 covered.

One week's sources in a list: two textbook reading notes and lecture recording notes, each marked ready — One week's material, grouped together: the readings and the lecture. From the demo account.

Study formats that work

Learning research is fairly clear on which methods pay off. A large review of ten common study techniques (Dunlosky and colleagues, 2013) rated practice testing and spreading study over time as the most useful. Rereading and highlighting rated low. AI is good at producing material for the methods that work:

Notes you can study from. Each idea on the slides explained in full, with the worked examples. See how to turn lecture slides into study notes.
Flashcards for definitions, formulas and facts you need to recall on demand.
Practice tests in the format of your real exam, graded with written feedback. See how to make a practice test from your course notes.
Questions about what you don't get. Ask the AI to explain a step, and ask it to point to the slide or page it used.

How to check what the AI writes

AI models make mistakes, and they make them with confidence. Three habits catch most of them:

Ask where each answer came from in your material, and look at that page.
Rework any calculation yourself. Treat the AI's numbers as a draft.
When the AI and your professor disagree, your professor wins. The exam follows the course.

A weekly routine

After each lecture, add that week's slides, readings and recording.
Read the notes the same week, while the lecture is fresh.
Do a short set of flashcards every few days, including older weeks.
Before each test, take a full practice test and study the questions you missed.

Spacing review across days beats cramming the same hours into one night (Cepeda and colleagues, 2006), so the routine matters more than any single long session.

General chatbot or course-aware tool?

A general chatbot is fine for one quick explanation. For a whole term of one class, a tool that keeps all your files and the syllabus does more of the work. The trade-offs are laid out in ChatGPT vs a study tool that knows your course. Sendy Study Tool is one tool built this way.

Sources

Dunlosky, J., Rawson, K. A., Marsh, E. J., Nathan, M. J., and Willingham, D. T. (2013). Improving students' learning with effective learning techniques. Psychological Science in the Public Interest, 14(1), 4-58. doi:10.1177/1529100612453266
Cepeda, N. J., Pashler, H., Vul, E., Wixted, J. T., and Rohrer, D. (2006). Distributed practice in verbal recall tasks. Psychological Bulletin, 132(3), 354-380. doi:10.1037/0033-2909.132.3.354

Questions students ask

Is it cheating to study with AI?

Studying with AI is not cheating when you use it to learn: making notes, quizzing yourself, or explaining a hard idea. Handing in AI-written work as your own usually is. Your school's academic integrity policy and your syllabus set the rules, so read both.

What is the best way to study with AI?

Give the AI your own course files, then use it to test you rather than to summarize for you. Practice questions and flashcards built from your material make you recall the content, and recall is what moves exam scores.

Can AI read my lecture slides?

Yes. Most current AI models read PDF and PowerPoint files, including the text on each slide. Some also read the diagrams. Slides alone are thin, so pair them with the reading or the lecture recording when you can.