Reference brief. This article explains published guidance and its practical context; it does not report a new launch or claim that an older source was released today.
Education AI should be judged by the learning it supports, not only the text it produces. Institutions evaluating a tutor should ask to see a wrong initial answer, a useful hint and an independent follow-up task. Those steps make the learner’s understanding more visible.
What the source establishes
UNESCO’s guidance on generative AI in education and research addresses human agency and responsible use. It offers policy and implementation context rather than proving a particular tutor’s effectiveness.
See the original reference material for the publisher’s scope and detail.
Questions to carry into a product or service review
- Prefer prompts that invite explanation and revision rather than answer copying.
- Review the accuracy and suitability of feedback for the learner’s level.
- Set clear expectations for permitted use and handling of learner information.
Why context matters
Helpful scaffolding and excessive assistance can look similar in a chat transcript. Teachers need a way to inspect the learning process and adapt support rather than relying on the tool’s confident tone.
For the practical example, read use an ai tutor to reveal a learner’s reasoning, not hide it. More coverage is available in Artificial Intelligence in Education.
References and further reading
- unesco.orghttps://www.unesco.org/en/articles/guidance-generative-ai-education-and-research
