Home · Sources · Oreopoulos et al. 2026
Making AI Tutoring Productive: Evidence from a Mastery-Based Math Practice Experiment
Oreopoulos et al. 2026 · randomised trial · touches 2 claims
Read the study itself (doi.org) ↗
Which beliefs it touches
- C2 Answer-giving AI in practice sets a limit on it
Guard-railed AI inside a mastery rule gave a modest delayed gain over the same platform without AI; not answer-giving, so outside C2.
- C5 Step-based tutoring sets a limit on it
AI tutor inside a mastery rule gave a modest delayed gain over the same platform without AI; a large-language-model tutor, no human-tutor comparison.