Your teaching material, turned into a conversation.
Built by learning researchers: every student gets a private tutor grounded in your syllabus, citing the page it answered from — and built on established didactics to make them reason rather than receive. Two steps to get a course running.
Several agents, each with a distinct way of explaining, become material for discussion in class.
- One agent per response style — short, detailed, step-by-step
- Students compare explanations of the same question
- Selected answers reused as classroom discussion material
Lecturer at Copenhagen Business School
Answers stay anchored in the uploaded course material, and question patterns feed back into the next lecture.
- Notes and articles uploaded, so answers match what is taught
- Analytics surface recurring misunderstandings across the group
- Lectures and exercises adjusted to the gaps that appear
Professor at Technical University of Denmark
ChatTutor plays the applicant company; students interrogate it from assigned professional roles.
- Full application dossier loaded; the agent answers as the company
- Students work as regulator, bioethicist, reimbursement specialist
- Findings compared in class: which questions exposed most
Associate Professor at University of Copenhagen
A shared session lets the whole room answer the same question at once — then @ai clusters the answers live.
- Every student answers, not just the few who speak up
- Typing @ai groups answers and surfaces misconceptions
- Contributions kept, so participation can be reviewed after class
Lecturer at Universiti Teknologi MARA (UiTM)
Two steps. Then your course is live.
Drop in your PDFs — slides, notes, problem sets. Our parsing handles equation-rich STEM material, so the agent can answer from the maths, not around it.
Share one course link or invite by email. Students are in immediately — no install, no accounts to provision, nothing for IT to approve first.
Built to make students think more, not less.
What people ask before they start.
What is ChatTutor?
ChatTutor is an AI tutoring platform for higher education courses. Teachers upload their course material, and every student gets a private tutor that answers only from that syllabus — citing the exact page it answered from.
How long does setup take?
Two steps. Uploading course material is typically under an hour of your time; students are enrolled with one course link the same day. Nothing to install, nothing for IT to approve first.
Can I try it without involving IT or procurement?
Yes — that is what the trial is for. One course, your own material, no credit card, nothing to install. Procurement only becomes relevant when you move beyond the trial to a department or faculty.
Do students need an account?
Nothing to set up on their side. You share one course link, they sign in with their university email, and they are in. No install, no separate password to remember.
What can I upload?
PDFs, slides, lecture notes, video, code and more. Text is extracted page by page, so answers can cite the exact source. A full semester reader is a normal upload, not an edge case.
What does it cost?
Trying it is free for one course, no credit card. After that, pricing follows course size and is quoted per institution — or buy a single course outright and invoice it to your department.
Does ChatTutor grade or assess students?
No. ChatTutor supports learning; assessment stays with the teacher. The boundary is deliberate — it is also what keeps the platform out of the EU AI Act's high-risk category.
Can the agent answer outside the course material?
It is grounded in what you upload and cites the page it answered from. You set how far it may go beyond that, per course.
What about students using it to cheat?
The agent is configured to coach rather than complete, guardrails are set per course, and teachers can read every conversation in their own course. Assessment integrity stays a teaching decision — with the evidence visible.
Which AI model does it use?
Whichever you choose. ChatTutor is model-agnostic: the model is selected per conversation and can be changed mid-thread, so neither you nor your institution is locked to one vendor.
Does it work in my language?
Students can ask and be answered in their own language while the course material stays in the language you uploaded it in. Useful for international cohorts and exchange students.
Is it compliant with GDPR, FERPA and the EU AI Act?
Yes — documented purpose, human oversight of teaching decisions, students told they are talking to an agent, education-records handling that meets FERPA, and a WCAG 2.1 AA interface. Full compliance page.
What happens at the end of the semester?
You decide. A course can be archived and reused with the next cohort, or deleted — which removes its indexed files and conversations. Retention periods are set with your institution in the data processing agreement.
Is ChatTutor free for researchers?
A pilot of up to 50 students for one semester is free. Larger studies are agreed with your institution, and we also join grant applications as a partner or co-applicant. Research collaboration.
Who is behind ChatTutor?
ChatTutor is built by HumanThinking, a research-focused company working with university faculty on how students learn alongside AI. Meet the team.
Pick the one that fits your semester.
No pricing pages to decode. Tell us the course size and we quote it — or skip us entirely and start on your own.