Newly announced

• AI Alignment Foundation (AIAF) (deadline: rolling)

  • Funds alignment researchers with ideas they can’t pursue within their current institutional funding and experts from other fields whose perspectives could produce breakthroughs.

  • Consider applying if: you’re proposing a new research direction rather than incremental work on mainstream agendas. Alongside funding, they provide dedicated engineering, compute, and infrastructure teams.

Closing soon

None.

Featured ongoing opportunities

We include these every time. See the directory for the full list.

• Coefficient Giving

  • Career development and transition funding: Supports people at any career stage pursuing careers focused on reducing global catastrophic risks – including graduate study, professional training, and self-study.

  • Capacity-building on transformative AI risks: Funds projects that support people working in AI safety, help people transition to the field, or contribute to public discourse about transformative AI.

• Transformative AI Fund (TAIF)

  • Funds new projects addressing transformative AI, with grants typically $10k–$150k. Run by Effective Altruism Funds. Replaced the Long-Term Future Fund (LTFF).

• Lightcone Commons

  • New grantmaking platform coordinating large-scale philanthropy in quarterly rounds. Simple application process.

• Manifund

  • Regranting platform where individual regrantors have funding pools to direct towards publicly-listed projects. Regrantors initiate fast, small grants, seeding early-stage projects with $5k–$50k.

Newly announced

• Transformative AI Fund (TAIF) (deadline: rolling)

  • Funds new projects addressing transformative AI, with grants typically $10k–$150k. Run by Effective Altruism Funds. Replaced the Long-Term Future Fund (LTFF).

  • Consider applying if: you’re an individual or team getting something new off the ground in technical AI safety, governance, forecasting, or field-building – including new initiatives within existing orgs. Speculative bets welcome.

Closing soon

• Lightcone Commons (deadline: 23 August)

  • New grantmaking platform coordinating large-scale philanthropy in quarterly rounds, expecting to direct $15–25M in this first round. Simple application process.

  • Consider applying if: you want many funders to see one application. No 501(c)3 status is necessary – individuals, for-profits, educational institutions, and non-US organizations are all welcome to apply.

• Corrigibility Research Fund (deadline: 23 August)

  • Awards grants and prizes for corrigibility research – aiming to ensure AI keeps humans informed and in control. Managed by Max Harms and housed at Lightcone Infrastructure.

  • Consider applying if: you need around $5k–$35k for corrigibility research that’s compatible with mainstream ML and useful to the people building AI systems. There’s no location or affiliation requirement.

Featured ongoing opportunities

We include these every time. See the directory for the full list.

• Coefficient Giving

  • Career development and transition funding: Supports people at any career stage pursuing careers focused on reducing global catastrophic risks – including graduate study, professional training, and self-study.

  • Capacity-building on transformative AI risks: Funds projects that support people working in AI safety, help people transition to the field, or contribute to public discourse about transformative AI.

• Manifund

  • Regranting platform where individual regrantors have funding pools to direct towards publicly-listed projects. Regrantors initiate fast, small grants, seeding early-stage projects with $5k–$50k.

Newly added

• Iliad: Request for Proposals (deadline: rolling)

  • Funds theory-driven research on ReLU networks and projects that raise the epistemic bar of AI safety. Mathematicians especially welcome, regardless of prior AI experience.

  • Consider applying if: you could build a new research org around one of these directions. The epistemic-bar track must be led by senior researchers with experience fixing epistemic problems.

• Iliad: Open Call for Research Projects (deadline: rolling)

  • Funds new research orgs and projects meeting a high standard of scientific rigor – predictive, mathematically founded theory and methodologically mature experiments.

  • Consider applying if: you have a concrete research bet with a clear path to safer AI, and prior work showing you can execute at that level. Both existing groups and founders seeking incubation are welcome.

Closing soon

• Advanced Research + Invention Agency (ARIA): Mathematics for Safe AI (deadline: 31 July)

  • UK government R&D funding agency aiming to unlock scientific and technological breakthroughs that benefit everyone. Similar to DARPA in the US.

  • Consider applying if: you have a bold idea fitting its Mathematics for Safe AI space. Seeds run £10k–£500k and unaffiliated individuals can apply (idea matters more than CV), but most work must happen in the UK.

• Schmidt Sciences: Multi-Agent Safety (deadline: 8 August)

  • Funds research studying the safety of large networks of interacting AI agents, run jointly with Google DeepMind, ARIA, and CAIF. Offering up to $1M over 1–2 years.

  • Consider applying if: you’re a technical researcher (anywhere in the world) studying risks from networks of interacting AI agents. Policy advocacy and commercial work are out of scope.

Featured ongoing opportunities

We include these every time. See the directory for the full list.

• Coefficient Giving

  • Career development and transition funding: Supports people at any career stage pursuing careers focused on reducing global catastrophic risks – including graduate study, professional training, and self-study.

  • Capacity-building on transformative AI risks: Funds projects that support people working in AI safety, help people transition to the field, or contribute to public discourse about transformative AI.

• Long-Term Future Fund (LTFF)

  • Makes small grants (generally <$100k) addressing global catastrophic risks. Historically one of the best funders for small projects and individuals, but has since become funding-constrained.

• Manifund

  • Regranting platform where individual regrantors have funding pools to direct towards publicly-listed projects. Regrantors initiate fast, small grants, seeding early-stage projects with $5k–$50k.

Newly announced/added

• Lightcone Commons (deadline: 23 August)

  • New grantmaking platform coordinating large-scale philanthropy in quarterly rounds, expecting to direct $15–25M in this first round. Simple application process.

  • Consider applying if: you want many funders to see one application. No 501(c)3 status is necessary – individuals, for-profits, educational institutions, and non-US organizations are all welcome to apply.

• Corrigibility Research Fund (deadline: 23 August)

  • Awards grants and prizes for corrigibility research – aiming to ensure AI keeps humans informed and in control. Managed by Max Harms and housed at Lightcone Infrastructure.

  • Consider applying if: you need around $5k–$35k for corrigibility research that’s compatible with mainstream ML and useful to the people building real systems. There’s no location or affiliation requirement.

• Paradigm 3 (deadline: rolling)

  • Funds research on AI externalities, evals science, alignment questions, slowing AI progress, and questions about AI capability trajectories.

  • Consider applying if: you have a concrete proposal in one of their focus areas. No application form, instead pitch the team by email.

Closing soon

• grantmaking.ai (deadline: 27 July)

  • Public database of AI safety projects seeking funding, currently running a $1M round offering $5K–$50K grants. Applications will also be visible to other funders.

  • Consider applying if: you want funding without heavy paperwork – the form takes ~10 minutes, individuals can apply (no legal entity needed), and career-change or exploratory work is considered.

• Advanced Research + Invention Agency (ARIA): Mathematics for Safe AI (deadline: 31 July)

  • UK government R&D funding agency aiming to unlock scientific and technological breakthroughs that benefit everyone. Similar to DARPA in the US.

  • Consider applying if: you have a bold idea fitting its Mathematics for Safe AI space. Seeds run £10k–£500k, unaffiliated individuals can apply (idea matters more than CV). Most work must happen in the UK.

Featured ongoing opportunities

We include these every time. See the directory for the full list.

• Coefficient Giving

  • Career development and transition funding: Supports people at any career stage pursuing careers focused on reducing global catastrophic risks – including graduate study, professional training, and self-study.

  • Capacity-building on transformative AI risks: Funds projects that support people working in AI safety, help people transition to the field, or contribute to public discourse about transformative AI.

• Long-Term Future Fund (LTFF)

  • Makes small grants (generally <$100k) addressing global catastrophic risks. Historically one of the best funders for small projects and individuals, but has since become funding-constrained.

• Manifund

  • Regranting platform where individual regrantors have funding pools to direct towards publicly-listed projects. Regrantors initiate fast, small grants, seeding early-stage projects with $5k–$50k.

Newly announced

• grantmaking.ai (deadline: 13 July for priority review, final deadline 27 July)

  • Public database of AI safety projects seeking funding, currently running a $1M round offering $5K–$50K grants. Applications will also be visible to other funders.

  • Consider applying if: you want funding without heavy paperwork – the form takes ~10 minutes, individuals can apply (no legal entity needed), and career-change or exploratory work is considered.

Closing soon

• Tarbell: AI Reporting Grants (deadline: 12 July)

  • Grants of $1k–20k to support journalism on AI and its impacts. Mainly focuses on written journalism, but also funds other formats.

  • Consider applying if: you’re an experienced journalist or freelancer who wants to report on AI’s risks and impacts – no prior AI-beat experience or secured outlet needed.

Featured ongoing opportunities

We include these every time. See the directory for the full list.

• Coefficient Giving

  • Career development and transition funding: Supports people at any career stage pursuing careers focused on reducing global catastrophic risks – including graduate study, professional training, and self-study.

  • Capacity-building on transformative AI risks: Funds projects that are supporting people working in AI safety, helping people transition to the field, or contributing to public discourse about transformative AI.

• Long-Term Future Fund (LTFF)

  • Makes small grants (generally <$100k) addressing global catastrophic risks. Historically one of the best funders for small projects and individuals, but has since become funding-constrained.

• Manifund

  • Regranting platform where individual regrantors have funding pools to direct towards publicly-listed projects. Regrantors initiate fast, small grants, seeding early-stage projects with $5k–$50k.

Newly announced

• Leo Gao’s Experimental Microgranting Program (deadline: rolling)

  • $10k grants for ambitious, unusual projects to make AGI go well, via a deliberately minimal application. Open to anyone, including those new to the field.

  • Consider applying if: you’ve held off other grants due to application complexity or feeling too junior, or simply need a small grant for work related to AI safety.

Closing soon

• Longview Philanthropy: Extreme Power Concentration RFP (deadline: 2 July)

  • Funds proposals to reduce risks of AI enabling extreme concentration of power. 2 tracks: grants ($100K–$2M/yr) for projects across 12 priority areas, and career transition funding.

  • Consider applying if: you’re a researcher, team, or new org working on AI-enabled power concentration (governance, oversight, surveillance, autonomous weapons), or an individual transitioning into the field.

Featured ongoing opportunities

We include these every time. See the directory for the full list.

• Coefficient Giving

  • Capacity-building on transformative AI risks: Funds projects that are supporting people working in AI safety, helping people transition to the field, or contributing to public discourse about transformative AI.

  • Career development and transition funding: Supports people at any career stage pursuing careers focused on reducing global catastrophic risks – including graduate study, professional training, and self-study.

• Long-Term Future Fund (LTFF)

  • Makes small grants (generally <$100k) addressing global catastrophic risks. Historically one of the best funders for small projects and individuals, but has recently become funding-constrained.

• Manifund

  • Regranting platform where individual regrantors have funding pools to direct towards publicly-listed projects. Regrantors initiate fast, small grants, seeding early-stage projects with $5k–$50k.

Newly announced

• Schmidt Sciences: Scaling AI Safety for a Multi-Agent World (deadline: 8 August)

  • Funds research studying the safety of large networks of interacting AI agents, run jointly with Google DeepMind, ARIA, and others. Offering up to $1M over 1–2 years.

  • Consider applying if: you’re a technical researcher (anywhere in the world) studying risks from networks of interacting AI agents. Policy advocacy and commercial work are out of scope.

Closing soon

• AI Forge (register interest by 22 June)

  • Joint DARPA and NSF program funding US university-led research on interpretability, control, and adversarial robustness for national security applications.

  • Consider applying if: you lead or work in a US university research group on interpretability, control, or adversarial robustness.

Featured ongoing opportunities

We include these every time. See the directory for the full list.

• Coefficient Giving

  • Capacity-building on transformative AI risks: Funds projects that are supporting people working in AI safety, helping people transition to the field, or contributing to public discourse about transformative AI.

  • Career development and transition funding: Supports people at any career stage pursuing careers focused on reducing global catastrophic risks – including graduate study, professional training, and self-study.

• Long-Term Future Fund (LTFF)

  • Makes small grants (generally <$100k) addressing global catastrophic risks. Historically one of the best funders for small projects and individuals, but has recently become funding-constrained.

• Manifund

  • Regranting platform where individual regrantors have funding pools to direct towards publicly-listed projects. Regrantors initiate fast, small grants, seeding early-stage projects with $5k–$50k.

Newly announced

• AI Forge (register interest by 22 June)

  • Joint DARPA and NSF program funding US university-led research on interpretability, control, and adversarial robustness for national security applications.

  • Consider applying if: you lead or work in a US university research group on interpretability, control, or adversarial robustness.

• Tarbell: AI Reporting Grants (deadline: 12 July)

  • Grants of $1k–20k to support journalism on AI and its impacts. Mainly focuses on written journalism, but also funds other formats.

  • Consider applying if: you’re an experienced journalist or media creator, whether staff or freelance — no prior AI-beat experience or secured outlet needed.

Closing soon

None.

Featured ongoing opportunities

We include these every time. See the directory for the full list.

• Coefficient Giving

  • Capacity-building on transformative AI risks: Funds projects that are supporting people working in AI safety, helping people transition to the field, or contributing to public discourse about transformative AI.

  • Career development and transition funding: Supports people at any career stage pursuing careers focused on reducing global catastrophic risks – including graduate study, professional training, and self-study.

• Long-Term Future Fund (LTFF)

  • Makes small grants (generally <$100k) addressing global catastrophic risks. Historically one of the best funders for small projects and individuals, but has recently become funding-constrained.

• Manifund

  • Regranting platform where individual regrantors have funding pools to direct towards publicly-listed projects. Regrantors initiate fast, small grants, seeding early-stage projects with $5k–$50k.

Newly announced

• Longview Philanthropy: Extreme Power Concentration RFP (deadline: 2 July)

  • Funds proposals to reduce risks of AI enabling extreme concentration of power. 2 tracks: grants ($100K–$2M/yr) for projects across 12 priority areas, and career transition funding.

  • Consider applying if: you’re a researcher, team, or new organization working on AI-enabled power concentration (governance, oversight, surveillance, autonomous weapons etc.), or an individual transitioning into the field.

Closing soon

None.

Featured ongoing opportunities

We include these every time. See the directory for the full list.

• Coefficient Giving

  • Capacity-building on transformative AI risks: Funds projects that are supporting people working in AI safety, helping people transition to the field, or contributing to public discourse about transformative AI.

  • Career development and transition funding: Supports people at any career stage pursuing careers focused on reducing global catastrophic risks – including graduate study, professional training, and self-study.

• Long-Term Future Fund (LTFF)

  • Makes small grants (generally <$100k) addressing global catastrophic risks. Historically one of the best funders for small projects and individuals, but has recently become funding-constrained.

• Manifund

  • Regranting platform where individual regrantors have funding pools to direct towards publicly-listed projects. Regrantors initiate fast, small grants, seeding early-stage projects with $5k–$50k.

Newly announced/added

None.

Closing soon

• Science of Trustworthy AI (deadline: 18 May)

  • Schmidt Sciences program making grants of up to $5M+ for technical research that improves our ability to understand, predict, and control risks from frontier AI systems.

  • Consider applying if: you’re an AI safety researcher or research team at a university, national lab, institute, or nonprofit research org working on alignment, interpretability, or oversight. Ambitious proposals ($1M–5M+ over 1–3 years) are preferred.

• AI Interpretability (deadline: 26 May)

  • Schmidt Sciences is requesting proposals for a pilot program in interpretability, seeking new methods for detecting and mitigating deceptive behaviors in LLMs. Offering $300K–$1M for 1–3 years.

  • Consider applying if: you’re an academic or nonprofit researcher applying interpretability methods to detect or prevent deception in LLMs. Open globally.

Featured ongoing opportunities

We include these every time. See the directory for the full list.

• Coefficient Giving

  • Capacity-building on transformative AI risks: Funds projects that are supporting people working in AI safety, helping people transition to the field, or contributing to public discourse about transformative AI.

  • Career development and transition funding: Supports people at any career stage pursuing careers focused on reducing global catastrophic risks – including graduate study, professional training, and self-study.

• Long-Term Future Fund (LTFF)

  • Makes small grants (generally <$100k) addressing global catastrophic risks. Historically one of the best funders for small projects and individuals, but has recently become funding-constrained.

• Manifund

  • Regranting platform where individual regrantors have funding pools to direct towards publicly-listed projects. Regrantors initiate fast, small grants, seeding early-stage projects with $5k–$50k.

Newly announced

• BlueDot Impact: Career Transition Grants (deadline: rolling)

  • 6-12 months of runway for people transitioning into AI safety work. Includes introductions to people in the field, regular check-ins, and a community of fellow grantees.

  • Consider applying if: you've completed a BlueDot course in the past and have somewhat of a plan for transitioning to full-time work in AI safety.

Closing soon

None.

Featured ongoing opportunities

We include these every time. See the directory for the full list.

• Coefficient Giving

  • Capacity-building on transformative AI risks: Funds projects that are supporting people working in AI safety, helping people transition to the field, or contributing to public discourse about transformative AI.

  • Career development and transition funding: Supports people at any career stage pursuing careers focused on reducing global catastrophic risks – including graduate study, professional training, and self-study.

• Long-Term Future Fund (LTFF)

  • Makes small grants (generally <$100k) addressing global catastrophic risks. Historically one of the best funders for small projects and individuals, but has recently become funding-constrained.

• Manifund

  • Regranting platform where individual regranters have funding pools to direct towards publicly-listed projects. Regrantors initiate fast, small grants, seeding early-stage projects with $5k–$50k.

Newly announced/added

• Institute for Progress (IFP): The Launch Sequence (deadline: rolling)

  • $10k honorarium plus funder matchmaking for short pitches for projects preparing for advanced AI – including in alignment, control, evals, and AI security.

  • Consider applying if: you have a project idea and want help developing a full plan plus introductions to funders and potential co-founders.

• BlueDot Impact: Rapid Grants (deadline: rolling)

  • Grants of up to $10k for work on AI safety research, field-building, talent acceleration, etc. Applicants must have participated in/facilitated a BlueDot course.

  • Consider applying if: you’re a BlueDot alumnus or facilitator and need a small, fast grant e.g. for compute, events, travel, or tooling needs.

Closing soon

• Survival and Flourishing Fund (SFF) (deadline: 22 April)

  • Funds organizations working on humanity’s long-term survival and flourishing. Speculation Grants are rolling, while the the full S-Process runs annually and is currently open.

  • Consider applying if: you’re a nonprofit (or US/UK/Canada/Australia for-profit) seeking mid-to-large-scale funding for basically any kind of work on AI safety.

Featured ongoing opportunities

We include these every time. See the directory for the full list.

• Coefficient Giving

  • Capacity-building on transformative AI risks: Funds projects that are supporting people working in AI safety, helping people transition to the field, or contributing to public discourse about transformative AI.

  • Career development and transition funding: Supports people at any career stage pursuing careers focused on reducing global catastrophic risks – including graduate study, professional training, and self-study.

• Long-Term Future Fund (LTFF)

  • Makes small grants (generally <$100k) addressing global catastrophic risks. Historically one of the best funders for small projects and individuals, but has recently become funding-constrained.

• Manifund

  • Regranting platform where individual regranters have funding pools to direct towards publicly-listed projects. Regrantors initiate fast, small grants, seeding early-stage projects with $5k–$50k.

Newly announced/added

• AI Interpretability (deadline: 26 May)

  • Schmidt Sciences is requesting proposals for a pilot program in interpretability, seeking new methods for detecting and mitigating deceptive behaviors in LLMs. Offering $300K–$1M for 1–3 years.

  • Consider applying if: you’re an academic or nonprofit researcher applying interpretability methods to detect or prevent deception in LLMs. Open globally.

Closing soon

None.

Featured ongoing opportunities

We include these every time. See the directory for the full list.

• Coefficient Giving

  • Capacity-building on transformative AI risks: Funds projects that are supporting people working in AI safety, helping people transition to the field, or contributing to public discourse about transformative AI.

  • Career development and transition funding: Supports people at any career stage pursuing careers focused on reducing global catastrophic risks – including graduate study, professional training, and self-study.

• Long-Term Future Fund (LTFF)

  • Makes small grants (generally <$100k) addressing global catastrophic risks. Historically one of the best funders for small projects and individuals, but has recently become funding-constrained.

• Manifund

  • Regranting platform where individual regranters have funding pools to direct towards publicly-listed projects. Regrantors initiate fast, small grants, seeding early-stage projects with $5k–$50k.

Newly announced/added

• Survival and Flourishing Fund (SFF) (deadline: 22 April)

  • Funds organizations working on humanity’s long-term survival and flourishing. The full S-Process runs annually and is now open – it’s their main grant-making process, with recommendations informed by a group of independent assessors.

  • Consider applying if: you’re a nonprofit or for-profit org (for-profits must be based in the US, UK, Canada, or Australia) doing basically any kind of work on AI safety.

• Alignment Foundation (deadline: rolling)

  • Funding neglected approaches to AI alignment through grants and fiscal sponsorship to independent researchers, educational programs, and engineers developing mech interp tools.

  • Consider applying if: your work fits into any of those 3 categories. Projects run 3–24 months, and individuals, teams, academic institutions, and policy institutes are all eligible.

Closing soon

• Tarbell AI Reporting Grants (deadline: 8 March)

  • Grants of $1–20k to support journalism on AI and its impacts. Mainly focuses on written journalism, but also funds other formats (e.g. podcasts, newsletters, short documentaries).

  • Consider applying if: you’re an experienced journalist who wants to report on AI. An investigative background is a plus.

Featured ongoing opportunities

We include these every time. See the directory for the full list.

• Coefficient Giving

  • Capacity-building on transformative AI risks: Funds projects that are supporting people working in AI safety, helping people transition to the field, or contributing to public discourse about transformative AI.

  • Career development and transition funding: Supports people at any career stage pursuing careers focused on reducing global catastrophic risks – including graduate study, professional training, and self-study.

• Long-Term Future Fund (LTFF)

  • Makes small grants (generally <$100k) addressing global catastrophic risks. Historically one of the best funders for small projects and individuals, but has recently become funding-constrained.

• Manifund

  • Regranting platform where individual regranters have funding pools to direct towards publicly-listed projects. Regrantors initiate fast, small grants, seeding early-stage projects with $5k–$50k.

Newly announced/added

• Science of Trustworthy AI (deadline: 17 May)

  • Schmidt Sciences program making grants of up to $5M for technical research that improves our ability to understand, predict, and control risks from frontier AI systems.

  • Consider applying if: you’re an AI safety researcher or research team at a university, national lab, institute, or nonprofit research org working on alignment, interpretability, or oversight. Ambitious proposals ($1M–5M+ over 1–3 years) are preferred.

• Omidyar Network: Tech Journalism Fund (deadline: rolling)

  • Provides $5k–25k project funding for journalists covering topics related to how technology and society intersect.

  • Consider applying if: you’re a US-based journalist working on a resource-intensive story about technology’s societal impact – especially topics like AI regulation and the people, orgs, and ideas that are shaping the industry.

Closing soon

None.

Featured ongoing opportunities

We include these every time. See the directory for the full list.

• Survival & Flourishing Fund (SFF): Speculation Grants

  • Expedited grants from individual grantors, funded by Jaan Tallinn. Decisions can be made within 1 week for time-sensitive requests. Receiving a Speculation Grant guarantees eligibility for the next S-Process round.

• Coefficient Giving

  • Capacity-building on transformative AI risks: Funds projects that are supporting people working in AI safety, helping people transition to the field, or contributing to public discourse about transformative AI.

  • Career development and transition funding: Supports people at any career stage pursuing careers focused on reducing global catastrophic risks – including graduate study, professional training, and self-study.

• Long-Term Future Fund (LTFF)

  • Makes small grants (generally <$100k) addressing global catastrophic risks. Historically one of the best funders for small projects and individuals, but has recently become funding-constrained.

• ARIA: Mathematics for Safe AI

  • £10k–500k for up to 3 years, funding technical research aimed at creating mathematically provable safety guarantees for AI systems. Applicants need to be UK-based.

New opportunities

• Tarbell Grants (deadline: 8 March)

  • Grants of $1–20k to support journalism on AI and its impacts. Mainly focuses on written journalism, but also funds other formats (e.g. podcasts, newsletters, short documentaries).

  • Consider applying if: you’re an experienced journalist who wants to report on AI. An investigative background is a plus.

Closing soon

• 5050 (deadline: 15 February)

  • 12–14 week incubation program run by Fifty Years, helping scientists and engineers build startups working on AI safety.

  • Consider applying if: you’re a scientist or engineer (PhD, postdoc, or industry) exploring whether to start an AI safety company. You don’t need to already have an idea.

Featured ongoing opportunities

We include these every time. See the directory for the full list.

• Survival & Flourishing Fund (SFF): Speculation Grants

  • Expedited grants from individual grantors, funded by Jaan Tallinn. Decisions can be made within 1 week for time-sensitive requests. Receiving a Speculation Grant guarantees eligibility for the next S-Process round.

• Coefficient Giving

  • Capacity-building on transformative AI risks: Funds projects that are supporting people working in AI safety, helping people transition to the field, or contributing to public discourse about transformative AI.

  • Career development and transition funding: Supports people at any career stage pursuing careers focused on reducing global catastrophic risks – including graduate study, professional training, and self-study.

• Long-Term Future Fund (LTFF)

  • Makes small grants (generally <$100k) addressing global catastrophic risks. Historically one of the best funders for small projects and individuals, but has recently become funding-constrained.

• ARIA: Mathematics for Safe AI

  • £10k–500k for up to 3 years, funding technical research aimed at creating mathematically provable safety guarantees for AI systems. Applicants need to be UK-based.

New opportunities

• Future of Life Institute (FLI): Digital Media Accelerator (deadline: rolling)

  • Supports digital content from creators raising awareness and understanding about ongoing AI developments and issues.

  • Consider applying if: you’re a content creator (YouTube, blog, podcast, etc.) with an existing audience and you want to cover AI safety topics – especially if you can explain complex issues in accessible ways to non-technical viewers.

• 5050 (deadline: 15 February)

  • 12–14 week incubation program run by Fifty Years, helping scientists and engineers build startups working on AI safety.

  • Consider applying if: you’re a scientist or engineer (PhD, postdoc, or industry) exploring whether to start an AI safety company. You don’t need to already have an idea.

Closing soon

• Coefficient Giving: AI Governance (deadline: 25 January)

  • Funds a broad range of governance work targeting catastrophic risk from AI, including research, training programs, general org support, and other projects – events, tools, databases, etc.

  • Consider applying if: you’re doing basically any project that helps reduce AI x-risk through governance. $200k+/year grants are preferred.

• Future of Life Institute (FLI): religious projects (deadline: 2 February)

  • Funds impact-oriented religious initiatives to avert the risks from rapid AI development, as well as religious projects promoting pro-human visions of the future.

  • Consider applying if: you work with a religious organization (seminary, denominational body, interfaith council, etc.) and want to mobilize them on AGI risks.

Featured ongoing opportunities

We include these every time. See AISafety.com/funding for the full list.

• Survival & Flourishing Fund (SFF): Speculation Grants

  • Expedited grants from individual grantors, funded by Jaan Tallinn. Decisions can be made within 1 week for time-sensitive requests. Receiving a Speculation Grant guarantees eligibility for the next S-Process round.

• Coefficient Giving

  • Capacity-building on transformative AI risks: Funds projects that are supporting people working in AI safety, helping people transition to the field, or contributing to public discourse about transformative AI.

  • Career development and transition funding: Supports people at any career stage pursuing careers focused on reducing global catastrophic risks – including graduate study, professional training, and self-study.

• Long-Term Future Fund (LTFF)

  • Makes small grants (generally <$100k) addressing global catastrophic risks. Historically one of the best funders for small projects and individuals, but has recently become funding-constrained.

• ARIA: Mathematics for Safe AI

  • £10k–500k for up to 3 years, funding technical research aimed at creating mathematically provable safety guarantees for AI systems. Applicants need to be UK-based.

Recently added funders

(Quite a lot!)

• Babuschkin Ventures

  • New VC fund by Igor Babuschkin (xAI co-founder). Backs AI safety research and startups building agentic systems.

• Nonlinear AI Safety Advocacy Grants

  • Grant program providing funding to those raising awareness about AI risks or advocating for a pause in AI development.

• Halcyon Futures

  • Identifying leaders from business, policy, and academia, and helping them take on ambitious projects in AI safety.

• The Dreamery Foundation

  • Private foundation of Elastic co-founder Steven Schuurman, providing €5M/year to various initiatives including AI safety.

• Astera Institute

  • Private foundation providing grants to projects working to make the future with advanced AI go well. Also invests in for-profits in order to redeploy capital to AI safety nonprofits.

• Auspicious Fund

  • VC investing in startups shaping the future of AI and humanity, committed to delivering financial returns while building a resilient and beneficial future.

• Meta Charity Funders

  • Network of donors funding charitable projects that work one level removed from direct impact, often cross-cutting between cause areas.

• Anthology Fund

  • $100M fund by Menlo Ventures and Anthropic backing AI startups from seed to Series A, including those working on “trust and safety tooling”.

• Laidir Foundation

  • Private foundation funding AI governance: policymaker education on capabilities, policy design across a range of future scenarios, and civil society coordination on AI risk.

• Global Technology Risk (GTR) Foundation

  • Berlin-based foundation of tech investor Jan Beckers, funding research in LLM auditing/evaluations, interpretability, and global AI governance.

• Metaplanet

  • Evergreen VC and fund by Jaan Tallinn backing tech that supports humanity’s long-term survival. Profits from the VC go towards funding AI safety nonprofits.

Closing soon

• Future of Life Institute (FLI): Technical Postdoctoral Fellowships

  • $80k/year stipend plus $10k expenses fund for postdocs working on technical AI existential safety at US, UK, or Canadian universities. Includes invitations to FLI community events.

  • Consider applying if: you’re a postdoc (or about to be) with a confirmed mentor willing to supervise x-risk research – interpretability, alignment formalization, deception detection, etc.

  • Deadline: 5 January 2026

Featured ongoing funding opportunities

We include these every time. See AISafety.com/funding for the full list.

• Survival & Flourishing Fund (SFF): Speculation Grants

  • Expedited grants from individual grantors, funded by Jaan Tallinn. Decisions can be made within 1 week for time-sensitive requests. Receiving a Speculation Grant guarantees eligibility for the next S-Process round.

  • Consider applying if: you need funding faster than the regular S-Process rounds (which runs ~annually) and your work relates to AI safety in any form – research, advocacy, field-building etc.

• Coefficient Giving

  • AI Governance: Supports work across technical AI governance, policy development, frontier company policy, international governance, law, and strategic analysis/threat modeling.

  • Capacity-building on transformative AI risks: Supports projects doing things like helping people find work in the field, contributing to public discourse, or supporting professional networks.

  • GCR career development/transition funding: Supports people at any career stage pursuing careers focused on reducing global catastrophic risks, including graduate study, professional training, and self-study.

  • GCR programs and events: Supports programs and events related to global catastrophic risks and other areas.

• Long-Term Future Fund (LTFF)

  • Makes small grants (generally <$100k) addressing global catastrophic risks. Historically one of the best funders for small projects and individuals, but recently become funding-constrained.

  • Consider applying if: you have a small project or individual grant need for technical AI safety research, policy work, field-building, or other work aimed at reducing existential risk.

• ARIA: Mathematics for Safe AI

  • £10k–500k for up to 3 years. Part of ARIA’s Safeguarded AI program, which aims to create mathematically provable safety guarantees for AI systems. Applicants need to be UK-based.

  • Consider applying if: you’re doing technical research on mathematical approaches to AI safety. Open to early-career researchers and non-traditional backgrounds.

• Foresight Institute: AI for Science & Safety Nodes

  • $10k–100k grants for AI cybersecurity, decentralized/cooperative AI, and AI for epistemics/forecasting. Includes office space in San Francisco or Berlin, compute access, and workshop invitations.

  • Consider applying if: you’re building projects in these areas, especially if you want to work from San Francisco or Berlin. They prioritize short AGI timelines (1–3 year milestones), high-risk/high-reward, and open source.

New funding opportunities

• Martian Interpretability Challenge

  • Martian (a for-profit) is distributing $1M for mechanistic interpretability research, with a focus on understanding how LLMs write code. Distributed as grants for promising directions and awards for completed work.

  • Consider applying if: you’re doing interpretability research, especially on code generation, benchmarks, or methods that generalize across models.

  • Deadline: rolling

• Seldon Lab: Batch 02

  • Up to $500k for early-stage startups building AI safety infrastructure as part of a 3-month incubator program in San Francisco with office space, investor intros, and customer connections.

  • Consider applying if: you’re a highly technical founder (ideally with a team) building in areas like compute controls, runtime interpretability, or agentic system assurance.

  • Deadline: rolling (program starts mid-January 2026)

Closing soon

• –

Featured ongoing funding opportunities

See AISafety.com/funding for the full list

• Survival & Flourishing Fund (SFF): Speculation Grants

  • Expedited grants from individual grantors, funded by Jaan Tallinn. Decisions can be made within 1 week for time-sensitive requests. Receiving a Speculation Grant guarantees eligibility for the next S-Process round.

  • Consider applying if: you need funding faster than the regular S-Process rounds (which runs ~annually) and your work relates to AI safety in any form – research, advocacy, field-building etc.

• Coefficient Giving

  • AI Governance: Supports work across technical AI governance, policy development, frontier company policy, international governance, law, and strategic analysis/threat modeling.

  • Capacity-building on transformative AI risks: Supports projects doing things like helping people find work in the field, contributing to public discourse, or supporting professional networks.

  • GCR career development/transition funding: Supports people at any career stage pursuing careers focused on reducing global catastrophic risks, including graduate study, professional training, and self-study.

  • GCR programs and events: Supports programs and events related to global catastrophic risks and other areas.

• Long-Term Future Fund (LTFF)

  • Makes small grants (generally <$100k) addressing global catastrophic risks. Historically one of the best funders for small projects and individuals, but recently become funding-constrained.

  • Consider applying if: you have a small project or individual grant need for technical AI safety research, policy work, field-building, or other work aimed at reducing existential risk.

• ARIA: Mathematics for Safe AI

  • £10k–500k for up to 3 years. Part of ARIA’s Safeguarded AI program, which aims to create mathematically provable safety guarantees for AI systems. Applicants need to be UK-based.

  • Consider applying if: you’re doing technical research on mathematical approaches to AI safety. Open to early-career researchers and non-traditional backgrounds.

• Foresight Institute: AI for Science & Safety Nodes

  • $10k–100k grants for AI cybersecurity, decentralized/cooperative AI, and AI for epistemics/forecasting. Includes office space in San Francisco or Berlin, compute access, and workshop invitations.

  • Consider applying if: you’re building projects in these areas, especially if you want to work from San Francisco or Berlin. They prioritize short AGI timelines (1–3 year milestones), high-risk/high-reward, and open source.