OpenAI Commits $5 Million to Research AI's Impact on Teen Development

OpenAI will spend $5 million on independent research into how generative AI affects young people aged 13–17, with individual grants of up to $1 million each. The company opened applications now and will accept them through October 6, 2026, with selected proposals notified on or before November 13, 2026. OpenAI Group PBC funds and administers the program.

The announcement lands as generative AI becomes part of how teens learn, create, and seek advice, while regulators and policymakers debate what safeguards should apply to minors. OpenAI frames its goal plainly: "We want to build a stronger evidence base about the way teens use AI, as well as the impacts of AI on their lives, to inform development of AI products for teens and to help answer major questions facing policymakers and regulators."

What will the $5 million fund?

The program targets adolescence, with a particular focus on teens' social and emotional development. OpenAI acknowledges the effects of AI on teens are "likely to be nuanced," depending on what teens use AI for, how they engage with it, their individual circumstances, and the support available from families, schools, communities, and technology providers.

The company welcomes proposals that deepen understanding of:

• How teens use generative AI and how different patterns of use relate to developmental outcomes
• Potential impacts of AI use in teens' lives
• Individual, social, cultural, and technical factors that shape those effects
• Technical interventions, safeguards, and design choices that increase benefits or keep teens safer
• How effects differ across ages, developmental stages, populations, cultures, languages, and socioeconomic contexts

OpenAI encourages interdisciplinary work spanning child and adolescent development, psychology, public health, human-computer interaction, communication, sociology, anthropology, data science, and AI safety. Proposals from any country qualify, with priority given to countries with significant AI adoption. Qualitative, quantitative, experimental, observational, theoretical, participatory, and mixed-methods approaches are all eligible.

What research topics get priority?

OpenAI lists four priority areas:

• Emotional development and well-being: emotion regulation, resilience, identity, agency, self-esteem, and social connection
• Social development and relationships: friendships, family relationships, belonging, communication, empathy, social skills, and the relationship between AI and human connection
• Patterns, contexts, and variance across demographics: how outcomes vary by time spent, type and purpose of use, age, culture, language, and socioeconomic circumstances
• Mitigations, age-appropriate design, and policy approaches: safeguards, product interventions, AI literacy, parental or educator support, and healthy-use practices

Preference may go to research involving ChatGPT, interdisciplinary teams, underrepresented populations or regions, and projects capable of producing clear evidence that can inform near- or medium-term policy and regulatory decisions.

How will OpenAI judge independence?

The review criteria include a notable clause on credibility. Proposals will be evaluated on "the project's ability to produce trustworthy findings regardless of whether those findings are favorable to providers of AI products." Other criteria cover scientific rigor, relevance, actionability, ethics and safeguarding, feasibility, team expertise, inclusion and generalizability, and complementarity with OpenAI's internal research agenda.

Applicants must disclose relevant financial relationships, industry funding, or other potential conflicts of interest. A panel of internal experts and advisors will review applications on a rolling basis.

What does OpenAI require for research involving minors?

Any study involving minors demands extra scrutiny. Applicants must explain their institutional or independent ethics-review process, how informed consent and age-appropriate assent will be handled, how participants' privacy and data will be protected, and how researchers will respond to disclosures of harm or imminent risk.

Applicants must also state whether the project involves sensitive personal, behavioral, or mental-health information, what child-development, clinical, or safeguarding expertise the team holds, and whether parental consent applies. Collecting sensitive information does not automatically disqualify a project, but may trigger additional ethical review, expertise requirements, safeguards, and time.

What are the application rules?

Applications must be formatted as a single-spaced Google doc and submitted in English, though the research itself may be conducted in any language. Required components:

• Project summary, under one page
• Research proposal, no more than three pages excluding references, covering research questions, prior research, methods, study population, recruitment, expected outputs, limitations, and relevance
• Project plan and timeline with milestones and completion date
• Budget and justification; overhead or indirect costs capped at 10% of each grant
• Project dependencies and risks
• Ethics and safeguarding statement where relevant
• Team information, including short CVs or bios
• Independence and conflicts disclosure

To be eligible, applicants must be 18 or older and affiliated with a research institution or organization, or have significant relevant experience. OpenAI will not prioritize for-profit organizations. Multiple institutions may collaborate on one proposal, and grant funds can support student, graduate student, or post-doctoral involvement.

What outputs and timeline does OpenAI expect?

Every funded project must deliver a final report describing the research question, methods, results, limitations, and recommendations for policymakers, regulators, technology providers, or other stakeholders where the findings support them. An interim update, potentially including early findings, is due in Q1 2027.

OpenAI strongly encourages researchers to publish findings through peer-reviewed publication, preprint, or public report, but publication is not a condition of funding. The company also states a strong preference for shorter timelines: a working draft suitable for discussion at a small conference is expected within 12 months of receiving funds.

Final grants may depend on executing a grant agreement, award letter, certification, or other documentation, and grantees must comply with those terms. Questions go to [email protected]; OpenAI will not provide status updates for applications that are not selected.

The program's structure — capped overhead, mandatory conflict disclosures, and an explicit independence criterion — signals OpenAI's awareness that funder-backed research on its own products will face scrutiny. Whether the resulting evidence shapes teen-facing AI design and regulation will depend on what the first cohort finds by 2027.