Deadline: 10-Sep-2025
Applications are now open for a new funding opportunity aimed at developing mitigations for safety and security risks from misaligned AI systems. This initiative, supported by The Alignment Fund, targets researchers and innovators working to address potential threats posed by advanced AI technologies.
Funding for projects typically ranges from £50,000 to £1,000,000, though proposals with higher funding needs may also be considered. Selected recipients will gain access to a package of resources, including up to £1 million in funding, dedicated compute access through AWS for large-scale experiments, and potential venture capital investment to help commercialise alignment solutions.
Funding packages may include financial support for salaries, research, and operational expenses. Compute credits will be provided for cloud-based resources needed for experimentation, training, and scaling. Additionally, recipients will receive support and guidance from a network of AI alignment experts and domain specialists to tailor their project resources based on goals and infrastructure needs.
Eligibility is open to researchers from academic institutions, non-profits, and for-profit organisations, subject to due diligence. Applicants must represent a legal entity registered in their country and eligible for funding through recognised research channels. The person named in the application must be authorised to apply on behalf of their organisation and agree to provide information for due diligence checks in compliance with GDPR.
The application process is divided into two stages. In Stage 1, applications will undergo initial screening for eligibility and due diligence. Eligible expressions of interest (EOIs) will then be reviewed against quality criteria. Applicants will be notified whether they advance to the next stage.
In Stage 2, shortlisted applicants will work with assigned Research Sponsors to refine and develop a full proposal. Final applications will be reviewed by an independent panel, with each funder scoring based on their own selection criteria. Final funding decisions will be made based on these assessments, and applicants will be informed of the results.
For more information, visit AI Security Institute.