A tutor that asks the next question instead of giving the answer
Put a model alongside a learner working through a problem so that it replies with a guiding question, a partial hint or a check on the step just taken, and holds back the finished solution until the learner has got there.
- Effort
- Weeks of work
- Skill level
- Some technical skill
- Organisation size
- Small business
- Value
- Quality, Time saved
Tools named for this
- A model prompted to withhold the solution and offer the next step
- A transcript the teacher or parent can read afterwards
- A human tutor or teacher the learner can reach when it stalls
What to check before you ship it in India
- Where the learner has not completed eighteen years, s.9(1) of the DPDP Act requires verifiable consent of a parent or lawful guardian before their personal data is processed at all — which for a conversational tutor means before the first message is stored.
- s.9(3) prohibits tracking, behavioural monitoring and targeted advertising directed at children. A tutor that profiles a child's struggles in order to sell a course to them is doing the thing this provision names, and there is no consent that cures it. The prohibition admits no consent defence — a parent cannot authorise it — but it is exemptible under s.9(4), and rule 12 with the Fourth Schedule of the DPDP Rules 2025 exercises that power for named classes such as educational institutions, purpose-bound and not inheritable by a vendor.
- The 2025 Rules relieve part of that duty only for NAMED CLASSES — a clinical establishment, an educational institution, a creche, a school-transport operator — and for a school only so far as tracking and behavioural monitoring serve that institution's own educational activities or the safety of children enrolled in it. There is no general education relief from s.9(1), and a consumer tutoring product is not one of the listed classes and does not inherit a school's exemption. That relief is written for the institution providing the service; a consumer tutoring product is not automatically standing inside it.
- The randomised trial most often cited for this pattern put the model beside a human tutor rather than in front of the student, and the tutors in it still reported suggestions that were not appropriate to the grade level being taught.
Sources
Every claim on this page traces to one of these, on the date it was read.
- The Digital Personal Data Protection Rules, 2025 (G.S.R. 846(E), 13 November 2025) — Fourth Schedule, Part A · Ministry of Electronics and Information Technology (Gazette of India text) · a rule · read 2026-09-01
- DPDP Rules, 2025 Notified — A Citizen-Centric Framework for Privacy Protection and Responsible Data Use · Press Information Bureau, Government of India · a rule · read 2026-09-01
- The Digital Personal Data Protection Act, 2023 (No. 22 of 2023) — most obligations commence 13 May 2027 under the DPDP Rules 2025 — s.9(3) · Ministry of Electronics and Information Technology · a rule · read 2026-09-01
- The Digital Personal Data Protection Act, 2023 (No. 22 of 2023) — most obligations commence 13 May 2027 under the DPDP Rules 2025 — s.9(3) · Ministry of Electronics and Information Technology · a rule · read 2026-09-01
- Tutor CoPilot: A Human-AI Approach for Scaling Real-Time Expertise (arXiv:2410.03017v2) · Wang, Ribeiro, Robinson, Loeb and Demszky (arXiv preprint, not peer-reviewed) · that this is done · read 2026-09-01
- Guidance for generative AI in education and research · UNESCO · that this is done · read 2026-09-01