A lot of the skepticism about AI in education is well earned. Nobody wants a system that grades essays it hasn't understood, or generates lesson plans a teacher then has to fix from scratch. The useful question isn't whether AI belongs in a classroom — it's which parts of a teacher's week were never about teaching in the first place.
The first draft, not the final word
Writing thirty individualized report card remarks at the end of a term is real work, but it isn't the part of teaching that requires a teacher's judgment — the marks and the knowledge of the student are. Gurubaa drafts the remark from the actual data; the teacher reads it, corrects it, and approves it. The judgment stays human. The typing doesn't have to.
Signal, not surveillance
The same logic applies to spotting a student who's quietly falling behind. A teacher managing five sections can't always catch the pattern of missed homework and dipping attendance across every student in real time — but a system watching the same data continuously can flag it early enough to matter, and hand the decision on what to do about it back to the teacher.