“to accelerate and expand the field of AI safety and security research”In its own words: frontiermodelforum.org
Fund of over $10M from Frontier Model Forum member companies and philanthropic partners for AI safety and security research; first run by the Meridian Institute, managed directly by the FMF since Meridian's 2025 closure.
Show detailsHide details
Pooled fund backed by Frontier Model Forum members and philanthropic partners, supporting independent AI safety and security research through themed RFPs (biosecurity, cybersecurity, agent evaluation) reviewed competitively. The remaining money goes to narrowly scoped projects on urgent bottlenecks.
Says it gives about $10M (2023)
Typical grant $250k ($250k–$455k)
On applying “The projects were selected from over 100 competitive proposals through a rigorous review process.” frontiermodelforum.org
Itemised here $7.5M in 22 grants
’24
’25
Led by Chris Meserole (Executive Director)Checked 29 Oct 2026
Money flows
Where its money came from and where it went. Pick years and what to show; hover a box or ribbon for details, click one to focus it, click the background to see everything.
How to apply
- Applications
- Closed
No open call described; remaining funds go to narrowly-scoped projects.
Biggest recipients
All grants
22 grants totalling $7.5M, newest first. Recipients link to their own pages; people who received grants without a public profile are shown by number.
2025 10 grants · $4.5M
| Date | Recipient | Amount | Purpose |
|---|---|---|---|
| 11 Dec 2025 | Apollo Research | ~$455k | AISF Round 2: Building black box scheming monitors for frontier AI agents. Amount undisclosed; share of >$5M across 11 grants. Amount here is an estimate: an equal share of the round's >$5M. source |
| 11 Dec 2025 | California Institute of Technology | ~$455k | AISF Round 2: AI-driven detection of protein mimetic biothreats with BioSentinel. Amount undisclosed; share of >$5M across 11 grants. Amount here is an estimate: an equal share of the round's >$5M. source |
| 11 Dec 2025 | Institute for Decentralized AI (Cosmos Institute) | ~$455k | AISF Round 2: Scalable, decentralized oversight for multi-agent networks. Amount undisclosed; share of >$5M across 11 grants. Amount here is an estimate: an equal share of the round's >$5M. source |
| 11 Dec 2025 | Faculty AI | ~$455k | AISF Round 2: Automated red-teaming for biosecurity risks. Amount undisclosed; share of >$5M across 11 grants. Amount here is an estimate: an equal share of the round's >$5M. source |
| 11 Dec 2025 | FAR.AI | ~$455k | AISF Round 2: Quantifying the safety-adversary gap in large language models. Amount undisclosed; share of >$5M across 11 grants. Amount here is an estimate: an equal share of the round's >$5M. source |
| 11 Dec 2025 | Morgan State University | ~$455k | AISF Round 2: Evaluating AI-assisted cybersecurity operations (human uplift with and without AI tools). Amount undisclosed; share of >$5M across 11 grants. Amount here is an estimate: an equal share of the round's >$5M. source |
| 11 Dec 2025 | Nemesys Insights LLC | ~$455k | AISF Round 2: ICS benchmark and human uplift study. Amount undisclosed; share of >$5M across 11 grants. Amount here is an estimate: an equal share of the round's >$5M. source |
| 11 Dec 2025 | SecureBio | ~$455k | AISF Round 2: Evaluations of agent AIs' execution of tasks that could enable large-scale harm. Amount undisclosed; share of >$5M across 11 grants. Amount here is an estimate: an equal share of the round's >$5M. source |
| 11 Dec 2025 | University of Illinois Urbana-Champaign | ~$455k | AISF Round 2: Cybersecurity risk evaluations of AI agents with computer interaction capabilities. Amount undisclosed; share of >$5M across 11 grants. Amount here is an estimate: an equal share of the round's >$5M. source |
| 11 Dec 2025 | University of Toronto | ~$455k | AISF Round 2: Emergent role of sanctioning in regulating multi-agent LLM systems. Amount undisclosed; share of >$5M across 11 grants. Amount here is an estimate: an equal share of the round's >$5M. source |
2024 12 grants · $3M
| Date | Recipient | Amount | Purpose |
|---|---|---|---|
| Nov 2024 | Nemesys Insights LLC | ~$250k | AISF Round 1 (PI Gary Ackerman): Biothreat Benchmark Generation (BBG) Framework. Amount undisclosed (grants were $150k–$400k); share of >$3M across 12 grants. Amount here is an estimate: an equal share of the round's >$3M. source |
| Nov 2024 | SecureBio | ~$250k | AISF Round 1 (PI Individual #8549): Virus Methods QA (VMQA) benchmark. Amount undisclosed (grants were $150k–$400k); share of >$3M across 12 grants. Amount here is an estimate: an equal share of the round's >$3M. source |
| Nov 2024 | AI Safety Cape Town | ~$250k | AISF Round 1 (PI Individual #9382): measuring threats to human agency posed by AI models. Amount undisclosed (grants were $150k–$400k); share of >$3M across 12 grants. Amount here is an estimate: an equal share of the round's >$3M. source |
| Nov 2024 | Center for Democracy & Technology (AI Governance Lab) | ~$250k | AISF Round 1 (PI Miranda Bogen, with Dylan Hadfield-Menell (MIT CSAIL)): effects of modifying foundation models on safety performance. Amount undisclosed (grants were $150k–$400k); share of >$3M across 12 grants. Amount here is an estimate: an equal share of the round's >$3M. source |
| Nov 2024 | Center for Security and Emerging Technology (CSET) | ~$250k | AISF Round 1 (PI Individual #6496): AI safety and security framework. Amount undisclosed (grants were $150k–$400k); share of >$3M across 12 grants. Amount here is an estimate: an equal share of the round's >$3M. source |
| Nov 2024 | ETH Zurich | ~$250k | AISF Round 1 (PI Florian Tramèr): data poisoning attacks that survive the alignment stage. Amount undisclosed (grants were $150k–$400k); share of >$3M across 12 grants. Amount here is an estimate: an equal share of the round's >$3M. source |
| Nov 2024 | Imperial College London | ~$250k | AISF Round 1 (PI Individual #6844): detecting and evaluating AI sandbagging. Amount undisclosed (grants were $150k–$400k); share of >$3M across 12 grants. Amount here is an estimate: an equal share of the round's >$3M. source |
| Nov 2024 | OpenMined Foundation | ~$250k | AISF Round 1 (PI Andrew Trask): technical AI governance infrastructure for structured model access. Amount undisclosed (grants were $150k–$400k); share of >$3M across 12 grants. Amount here is an estimate: an equal share of the round's >$3M. source |
| Nov 2024 | Stanford University | ~$250k | AISF Round 1 (PI Percy Liang): Foundation Model Transparency Index (FMTI). Amount undisclosed (grants were $150k–$400k); share of >$3M across 12 grants. Amount here is an estimate: an equal share of the round's >$3M. source |
| Nov 2024 | Virtue AI | ~$250k | AISF Round 1 (PI Bo Li, with Dawn Song): red-teaming strategies and toolbox for LLM agents. Amount undisclosed (grants were $150k–$400k); share of >$3M across 12 grants. Amount here is an estimate: an equal share of the round's >$3M. source |
| Nov 2024 | Apollo Research | ~$250k | AISF Round 1 (PI Marius Hobbhahn): white-box evaluations for AI deception. Amount undisclosed (grants were $150k–$400k); share of >$3M across 12 grants. Amount here is an estimate: an equal share of the round's >$3M. source |
| Nov 2024 | New York University | ~$250k | AISF Round 1 (PI Individual #3270): deceptive behaviour arising from reward hacking. Amount undisclosed (grants were $150k–$400k); share of >$3M across 12 grants. Amount here is an estimate: an equal share of the round's >$3M. source |