Mentored AI research offers a concrete high-school pathway
What would meaningful AI experience look like for high-school students beyond using a chatbot? EDIT AI offers virtual summer research at the intersection of computing and cancer medicine, supported by mentors and three tracks: skills development, advanced research and peer mentoring. Its 2020-2025 internal evaluation reports 284 participants from 24 states and 39 peer-reviewed publications. Students reported greater familiarity with biomedical AI; alumni also returned as near-peer mentors. Seminars, research teams and a final symposium formed part of the program. These results support the feasibility of sustained mentored participation, but the evaluation cannot establish that the program caused later academic or career success.
Why it matters for leaders
Explore one university partnership with protected mentor time, accessible computing and transparent recruitment. Require students to explain their research decisions and show completed work. Evaluate who can participate, including students needing equipment or schedule support, before treating publication counts as a measure of program value.
The evidence
The accepted-article abstract describes familiarity surveys, attendance, participant feedback and scholarly outputs. Detailed full-text methods were unavailable on the landing page.
Who was studied
High-school students in virtual biomedical AI research.
What the evidence cannot tell us
Internal evaluation and self-reports do not establish causal impact or equitable access. Publications are collective outputs, not evidence that every student benefited.
Read the original sources
Summary and leadership analysis from the Week ending 2026-09-18 brief. The edition’s selection notes explain source access and evidence limitations.