Use of Generative AI for Substance-Related Information Among Young Adults (Aged 18-29 Years) in the United States: Cross-Sectional Survey

Key Findings

• — 32.3% (95% CI 29.3%-35.2%; 321/995) reported substance-related generative AI use; 97.3% (968/995) reported any prior generative AI use
• — 57.5% (572/995) reported substance-related Google searches
• — Black respondents (aOR 2.52, 95% CI 1.72-3.7 vs white) and male respondents (aOR 2.04, 95% CI 1.55-2.7 vs female) were more likely to report substance-related generative AI use, adjusting for age and student status
• — Among substance-related users, 48.2% used it at least monthly; 92.8% (298/321) sought information on the consequences of use, 48.6% substance-specific information and 34.9% SUD-related topics
• — Perceived trustworthiness was rated higher for substance-related use (mean 3.66) than for general use (mean 3.51); perceived helpfulness was lower for substance-related use (4.12 vs 4.44)

Methodology Notes

Online nonprobability Prolific panel with quotas for age, sex and race (no Hispanic ethnicity field). Target n=1,000; final N=995 after 5 quality exclusions; 27 never-users coded as non-users. Newly adapted, unpiloted and unvalidated survey items; chi-square tests, one multivariable logistic regression, paired t tests. The paper reports 92.8% (298/321) both for the share of generative AI users who also used Google for substance queries and for the share seeking consequences-of-use information; the coincidence is in the source text. Submitted 2026-05-07, accepted 2026-09-14, published 2026-10-19 (JMIR Form Res 2026;10:e100667). Verified from the full text read through r.jina.ai on 2026-10-20 (JMIR direct returns a 202 WAF page).

Sources