top of page

Safe Learning-Enabled Systems




March 21, 2023



January 16, 2024




Award Floor:



Match Required?





Entity Types:

Nonprofits (with 501(c)(3) status), Public & State controlled institutions of higher education

*Who May Submit Proposals: Proposals may only be submitted by the following: -Non-profit, non-academic organizations: Independent museums, observatories, research laboratories, professional societies and similar organizations located in the U.S. that are directly associated with educational or research activities. -Institutions of Higher Education (IHEs) - Two- and four-year IHEs (including community colleges) accredited in, and having a campus located in the US, acting on behalf of their faculty members.Special Instructions for International Branch Campuses of US IHEs: If the proposal includes funding to be provided to an international branch campus of a US institution of higher education (including through use of subawards and consultant arrangements), the proposer must explain the benefit(s) to the project of performance at the international branch campus, and justify why the project activities cannot be performed at the US campus. *Who May Serve as PI: Employees of Open Philanthropy and Good Ventures may not participate in proposals submitted to this initiative, including as unfunded collaborators, via letters of collaboration or support, or via any other means.




Source Type:


As artificial intelligence (AI) systems rapidly increase in size, acquire new capabilities, and are deployed in high-stakes settings, their safety becomes extremely important. Ensuring system safety requires more than improving accuracy, efficiency, and scalability: it requires ensuring that systems are robust to extreme events, and monitoring them for anomalous and unsafe behavior. The objective of the Safe Learning-Enabled Systems program, which is a partnership between the National Science Foundation, Open Philanthropy and Good Ventures, is to foster foundational research that leads to the design and implementation of learning-enabled systems in which safety is ensured with high levels of confidence. While traditional machine learning systems are evaluated pointwise with respect to a fixed test set, such static coverage provides only limited assurance when exposed to unprecedented conditions in high-stakes operating environments. Verifying that learning components of such systems achieve safety guarantees for all possible inputs may be difficult, if not impossible. Instead, a systems safety guarantees will often need to be established with respect to systematically generated data from realistic (yet appropriately pessimistic) operating environments. Safety also requires resilience to unknown unknowns, which necessitates improved methods for monitoring for unexpected environmental hazards or anomalous system behaviors, including during deployment. In some instances, safety may further require new methods for reverse-engineering, inspecting, and interpreting the internal logic of learned models to identify unexpected behavior that could not be found by black-box testing alone, and methods for improving the performance by directly adapting the systems internal logic. Whatever the setting, any learning-enabled systems end-to-end safety guarantees must be specified clearly and precisely. Any system claiming to satisfy a safety specification must provide rigorous evidence, through analysis corroborated empirically and/or with mathematical proof.

Tired of searching for grants? Consider a Scouting Report.

bottom of page