Applications are open for the  Foresight Institute’s AI Safety Grant Program that offers funding for high-impact, high-risk research that directly tackles challenges in advanced AI safety. Projects must align with one or more of the program’s four focus areas:

Funding Focus Areas:

1. Security Technologies for AI Systems
   Scalable formal verification, automated red-teaming, decentralized compute infrastructure, and more.

2. Neurotechnology to Enhance or Compete with AGI
   Brain-aligned models, secure brain-computer interfaces, biologically inspired AI architectures.

3. Safe Multi-Agent Scenarios
   Detecting AI collusion or deception, building pro-social AI agents, simulation environments for AI-human interactions.

4. Automating Research and Forecasting
   AI tools for literature reviews, hypothesis generation, forecasting platforms, and collaborative decision-making systems.

Important: Proposals closely tied to active Requests for Proposals (RFPs) are strongly prioritized. General proposals are considered but are subject to a higher selection standard.

Media & Publicity:

Grantees may be featured on Foresight’s website, newsletter, and social media—with consent. Opt-out and confidentiality options are available.

Ineligible Proposals:

• General AI alignment work

• Sector-specific AI applications (e.g., AI for healthcare, education)

• Projects outside the defined safety focus areas

Deadline: Quarterly (March 31, June 30, September 30, December 31)