Studying with AI
without outsourcing your thinking.

Where AI really helps

• Explanations: a term in simple words, with an example or from a different angle.
• Feedback on your answer: write your own answer first, then have it checked for what is missing.
• Following the reasoning: not just which option is right, but why the others are wrong.
• Generating practice questions: have it create new questions on a chapter to practise active recall.

Where AI does the learning for you, in a bad way

If you only copy answers, you mostly learn how to operate an AI. Reading a good explanation quickly gives you the feeling of having understood. Whether it has really stuck only shows when you can answer the question without help. So the same rule applies as with any solution: think first, then check. More in the guide to active recall.

AI makes mistakes

Language models state wrong things convincingly too. So check important answers against a reliable source, ideally your own notes. An answer that matches your lecture is usually worth more for the exam than one that is right in general.

Rules and privacy

• Exam rules: in exams, your university's or provider's rules apply. Where AI tools are not allowed, don't use them.
• Essays and theses: many universities require you to declare AI use. Read your department's guidelines.
• Personal data: don't upload documents containing other people's data, such as patient cases or name lists.
• Copyright: your lecturers' notes are meant for your studies. Don't pass them on.

How KnowFi puts this into practice

KnowFi is built for the practice phase: you answer a question yourself, then KnowFi shows the answer and reasoning from your own notes. The reasoning comes first, not just the letter. What happens to your files is explained on Privacy and your notes.