THE HUMAN YIELD
SECTION 15
ISSUE 001
Make Socratic Tutoring the Default
Inference: begin tutoring with diagnosis, questions, and hints; reveal the answer late. The cited classroom trial improved perceived utility, not broad learning outcomes, so fluency cannot stand in for evidence. The real score is delayed understanding: can the student explain, transfer, and regulate the skill after the tutor leaves?
Why this idea is here
What the evidence establishes.
The OECD synthesizes the case for purposefully designed educational AI; a 2026 randomized classroom trial found that utility-value prompts improved perceived utility relative to a strategy condition, but neither intervention clearly improved effort, domain knowledge, or elaboration-based strategy use over standard ChatGPT.
Source ledger
Read the sources.
- S01OECD Digital Education Outlook 2026
OECD report / dated 2026 / retrieved 2026-07-09
- S02Enhancing School Students’ Self-Regulated Learning Through Generative AI Support
peer-reviewed randomized trial / dated 2026 / retrieved 2026-07-09