AI glossary · Beginner · Language models · Agents · Education
AI tutor
An AI tutor is an AI private tutor that answers each student's questions and adapts explanations and problems to their level. It is usually built on an LLM and supports learning through conversation.
About this term
- Level
- Beginner · Core ideas for first-time readers
- Category
- Industry AI
- Fields
- Language models, Agents
- Industries
- Education
- Added
- 2026-10-10
How it works
When a student asks a question or solves a problem, the AI looks at the answer, guesses where the student got stuck, and continues the explanation. A well-designed tutor guides students to think for themselves by giving hints or asking questions back rather than giving the answer right away. It may also adjust the difficulty of the next problem based on the student's work history. If textbooks or lecture materials are loaded in, it can be made to explain in line with that content.
The AI tutor learning loop
Until the student understands
- 1Student questions and work
- 2Guess where they're stuck
- 3Hints and follow-up questions
- 4Adjust difficulty
- 3. Hints and follow-up questions Prompt thinking instead of giving answers
- 4. Adjust difficulty Reflect work history
Why it matters
It is hard for one teacher in a classroom to give tailored explanations to every student. An AI tutor can provide help matched to each student's pace and level at any time, widening access for students who cannot easily get one-on-one tutoring. Teachers can use the AI's summary of each student's difficulties to decide who needs their direct help.
Things to know
If the AI writes homework answers for students, it can actually block learning, so designing it around hints is important. Because hallucinations can teach wrong concepts, it should be grounded in educational materials and checked by teachers. Students' learning records and conversations, especially those of minors, are sensitive personal information and require strict standards for storage and use. Caring for students' motivation and emotions remains the role of teachers and parents.
Example
Suppose a middle school student gets a linear equation problem wrong. Instead of showing the answer, the AI tutor asks, "What happens if you add the same number to both sides?" and, if it notices the student often gets the sign wrong when moving terms across the equals sign, gives a few more similar practice problems. After class, the teacher receives a summary of which steps students in the class struggled with most and prepares the next lesson.