AI Tutoring with Khanmigo in a Two-Year School Experiment
3 hours ago
- A two-year cluster randomized trial in 18 Tennessee middle schools tested Khan Academy's AI tutor, Khanmigo, configured to coach rather than give answers during remedial math sessions.
- Assignment to use Khanmigo raised math achievement by 1.3 national percentile ranks per term, or about 0.06 to 0.08 standard deviations over a school year, with full-year active participation yielding 0.14 standard deviations.
- These gains were similar to those from Khan Academy practice without AI assistance, suggesting the AI tutor did not outperform traditional digital practice.
- Student engagement with the tutor was low: 96% tried it, but the median student used it only a third of practice days and in only 17% of sessions after a mistake; most messages were bare answers or prompted clicks.
- The key challenge is getting students to actively use the AI tutor, not merely providing access, to realize its potential.