[Schmidt Sciences](https://www.schmidtsciences.org)

May 17, 2026 (Full proposal/application deadline)

Tier 1: Up to $1M (1–3 years) | Tier 2: $1M–5M+ (1–3 years)

Supports technical research to advance understanding, evaluation, and control of risks from advanced AI systems for trustworthy deployment.

Preference for ambitious, collaborative, field-shaping proposals; global eligibility; indirect costs capped at 10%.

--- ## Description The Science of Trustworthy AI program by Schmidt Sciences funds technical research aimed at understanding, predicting, and controlling risks from frontier AI systems to enable their trustworthy deployment. The initiative is guided by a detailed research agenda and seeks proposals that address one or more of its three main aims: (1) characterizing and forecasting misalignment in advanced AI systems, (2) developing generalizable measurements and interventions for AI evaluation and control, and (3) overseeing AI systems with superhuman capabilities and addressing multi-agent risks. The program encourages deep, focused research—especially collaborative projects—and offers not only funding but also access to computing resources, software engineering support, API credits, and community engagement opportunities. --- ## Due Dates - May 17, 2026: Full proposal/application deadline (by 11:59 PM AoE/EDT) --- ## Funding Amount - **Tier 1:** Up to $1,000,000 for 1–3 years - **Tier 2:** $1,000,000–$5,000,000+ for 1–3 years Applicants may submit proposals to either or both tiers and may submit multiple proposals. Projects at both tiers are welcome, but the program especially encourages ambitious, field-shaping Tier 2 proposals. --- ## Eligibility - Open to individual researchers, research teams, research institutions, and multi-institution collaborations. - Eligible organizations include universities (public/private, US and non-US), national laboratories, institutes, and non-profit research organizations. - For-profit entities may not lead projects but may participate as partners in collaborations led by eligible non-profit organizations. - Global eligibility: international institutions are welcome and may serve as lead applicants, subject to charitable status verification. - Indirect costs are capped at 10% of the total project budget. - Cross-institution and cross-country collaborations are encouraged but not required. - Junior faculty and early-career researchers are eligible. --- ## Application Process - Review the [Research Agenda](https://www.schmidtsciences.org/trustworthy-ai-research-agenda/) to ensure your proposal aligns with the program’s aims and priorities. - Prepare a focused proposal that addresses core scientific questions relevant to trustworthy AI, with clear objectives, methods, and expected outcomes. - Proposals should be technically rigorous, demonstrate potential for significant impact, and be feasible within the proposed budget and timeline. - Submit applications through the [online portal](https://schmidtsciences.smapply.io/prog/science_of_trustworthy_ai_rfp_2026/). - Multiple proposals per applicant/team are permitted. - Notification of decisions is expected in Summer 2026. --- ## Additional Information - Preference is given to collaborative proposals, especially those involving multiple PIs or labs. - Projects may request funding for compute or access to Schmidt Sciences’ computing resources (GPUs, CPUs, storage, and networking). - Additional support includes software engineering, API credits, and opportunities for convenings and workshops. - Out of scope: projects focused solely on interpretability (see [AI Interpretability program](https://www.schmidtsciences.org/ai-interpretability/)), jailbreak discovery, generic capability evaluations, near-term product engineering, fairness/bias research not tied to the agenda, policy/AI governance, epistemic integrity, and direct CBRN/ARA dangerous capability evaluations. - Indirect costs must not exceed 10%. - Awards are typically structured as annual gifts or grants, with future years contingent on progress. - For international recipients, equivalency determination may be required. --- ## External Links - [Program Homepage & Application Portal](https://schmidtsciences.smapply.io/prog/science_of_trustworthy_ai_rfp_2026/) - [Research Agenda](https://www.schmidtsciences.org/trustworthy-ai-research-agenda/) - [FAQ Document](https://docs.google.com/document/d/1nuqku7P9yQaxLPkeIhs10-21SEX77LwxexRwVmqSxkI/edit?usp=sharing) - [Virtual Institute for Scientific Software](https://www.schmidtsciences.org/viss/) - [AI Interpretability Program](https://www.schmidtsciences.org/ai-interpretability/) --- ## Contact Information | Purpose | Contact | |-----------------------|-----------------------------------------| | General inquiries | trustworthyai@schmidtsciences.org | | FAQ | [Trustworthy AI FAQ](https://docs.google.com/document/d/1nuqku7P9yQaxLPkeIhs10-21SEX77LwxexRwVmqSxkI/edit?usp=sharing) | ---

2026 Science of Trustworthy AI RFP

Sign in to see full grant listing