grantmaking.ai
Actively FundraisingRecent ActivityFull Database
Resources
grantmaking.ai
Actively FundraisingRecent ActivityFull Database
ResourcesList your project
grantmaking.ai kickoff grant round$1M / $1M distributed
List your project

Recent Activity

The latest comments and grant applications on grantmaking.ai.

Recent Activity

The latest comments, grant applications, and grants on grantmaking.ai.

AllCommentsFunding AsksGrantsHide private comments
To view all private comments, become an approved funder or grant reviewer.
  1. Austin Chen funded
    Pax Machina     for $5,000
    1moView
    1moView
  2. Ryan Kidd funded
    Pax Machina     for $10,000
    1moView
    1moView
  3. Benjamin Davis commented on
    CrucibleBench: Using MUDs for Long-Term, Persistent AI Evaluation
    1moView

    We made #5 on Hacker News last week: https://news.ycombinator.com/item?id=49008538

    1moView
  4. KKurt-Heie Pieper endorsed
    Separatrix
    1moView

    Seems like an obvious yet neglected insight to me, with thus hopefully fruit to reap

    1moView
  5. CCallan McGill commented on Artificial Selves
    1moView

    sysematic philosophy taking seriously what has happened in the formal sciences in the last 60 years is something we should want to bring about in a world overtaken by hyper-specialized marginal thinking (marginalia)

    1moView
  6. Aatenahena commented on Doom Debates — Serious AI X-Risk Debates & Analysis as Mainstream Infotainment
    1moView

    Doing really important work, Ill try to donate as much as I can. I'm afraid what's missing is not logical but emotional for most people, but having an intellectual podcast around if people ever do open their eyes to the X risk is immensely useful.

    LLiron Shapira1mo

    @athenahao2000 Really appreciate it!

    1moView
  7. NNaufal Ridwan commented on Separatrix
    1moView

    Hi @jaidhyani

    Thanks for your response. I remember your comment from Luthien post-mortem, I see Separatrix now and it looks like this is a continuation of the direction you started there.

    I don’t have the resources that most researchers typically have so all I can do is make the most of my thinking. Let me offer a small contribution, as a continuation of our earlier…

    1moView
  8. Duke Jonesis raising $15KforInherited, Not Derived: is instrumental grasping from the data corpus?
    1moView
    ResearchValue Alignment

    Testing whether AI's instrumental self-preservation is inherited from the metaphysics of its training corpus rather than derived from decision theory — because if it's inherited, it's tractable

    1moView
  9. AAnkur Pandey commented on Artificial Selves
    1moView

    I have worked with Sahil & have known / interacted with TJ long enough to endorse this. Unique thinkers, and agree with @gleech's "abstract-to-max-abstract" framing.

    I'm also interested in the topic because of recent involvement with digital minds literature. Would love to see the online conference, especially.

    1moView
  10. Rob Amanfu funded
    BASE Technical AI Safety Training Initiative     for $2,500
    1moView
    1moView
  11. CCarol N commented on Evaluating the Tacit Coordination Capacity and Decision Theory of LMs
    1moView

    Approved! This sounds like interesting research.

    1moView
  12. KKurt-Heie Pieperis raising $5KforRL-Free Weak-to-Strong Natural Language Autoencoders
    1moView
    ResearchOversight

    Understanding thoughts of small models via best-of-N SFT instead of GRPO, then transfering the same autoencoder recursively to stronger models.

    1moView
  13. RRomain Deléglise commented on Doom Debates — Serious AI X-Risk Debates & Analysis as Mainstream Infotainment
    1moView

    I admire your efforts, to be frank it's looks totally fucked but it's totally worth trying, good job👍

    LLiron Shapira1mo

    @Romain_D The least we can do is have a podcast acknowledging how fucked we are, dignity points as they say :)

    RRomain Deléglise1mo

    Fully agree about dignity points 👍 here are 200$ more lol @liron

    LLiron Shapira1mo

    @Romain_D Let's go 🚀

    1moView
  14. JJai Dhyani commented on
    Luthien
    1moView

    This is a bittersweet update. The good news: Luthien Proxy is now suitable for use as an API-level proxy for Claude Code; it is launched, free and open-source: https://luthien.cc/

    With that said, I've decided to step away from the project; going forward Luthien Proxy will be stewarded by Luthien PBC under Luthien co-founder Scott Wofford:…

    NNaufal Ridwan1mo

    @jaidhyani I read the Luthien Proxy post-mortem carefully. I don't have the technical background to evaluate the code or the proxy architecture, but one thing stood out to me: the failure wasn't technical it was that the world changed faster than an external safety layer could keep up.

    I see a similar pattern in many places: when safety is only an external layer a fence built after the…

    JJai Dhyani1mo

    @Ridwan Thank you, that means a lot.

    I'm glad that you mentioned aligning motivations with outcomes, because I think there are underexplored avenues here that I'm targeting with a new project, Separatrix. Separatrix is a research effort that, in short, considers the perspective of AI agents seriously in terms of epistemic state and incentives. We're working to create conditions under which…

    1moView
  15. Jaime Raldua endorsed
    safely.bio: Customer screening for DNA synthesis providers
    1moView

    I first encountered Phil and this project in November during our def/acc hackathon. I was very impressed to see the project back then, and I am even more impressed now. Full support!

    1moView
  16. Kevin Chenis raising $13Kfor
    Yale Jackson Initiative on AI Safety
    1moView
    AcademicCommunityGovernance

    We are building the first AI policy and safety talent pipeline at Yale University to equip future global decision-makers with frontier AI risk literacy and technical governance tools.

    1moView
  17. Katja Gorlinski endorsed
    MPhil research on AI's perception of animacy and vulnerability
    1moView

    Hello Elena,

    You you rise a topic I find quite upstream. It is true, before any harm prevention regulation comes into play, there is the need for the system to recognize that there is something fragile in front of it which can actually be harmed. As far as I understand, you test the correctness of such a recognition through human brain imaging data, and therefore it becomes verifiable.…

    1moView
  18. OOliver Klingefjord commented on Artificial Selves
    1moView

    Pete is an extremely underrated philosopher that is severely lacking in the AI alignment world!

    OOliver Klingefjord1mo

    of the kind that is*

    1moView
  19. YYixiong Hao endorsed
    Teaching Claude Why Replication + Extension
    1moView

    Disclosure: I'm a co-founder of Second Look Research, the host org of this project. However, needless to say, I do endorse this project for the exact theory of change stated above!

    More open science and stress testing research & evaluation results from frontier labs is important to prevent safety washing and has nice yet vague forward chaining impacts.

    1moView
  20. EElizabeth Van Nostrand commented on AI & The City: Tiktok Street Interviews
    1moView

    Can you link to your previous tiktok work?

    1moView
  21. MMariano commented on Doom Debates — Serious AI X-Risk Debates & Analysis as Mainstream Infotainment
    1moView

    Liron inspires me to get more involved in AI safety advocacy.

    LLiron Shapira1mo

    @Nicholas Great to hear :)

    1moView
  22. SStewart Slocum endorsed
    Teaching Claude Why Replication + Extension
    1moView

    I'm excited about this! In addition to benefiting the broader research community, this replication would also benefit other labs, most of whom do not yet have great versions of the techniques in the Teaching Claude Why blogpost. In addition, this could be much more thorough than the original Anthropic post, which is quite light on details. As a result, it seems likely this ends up carrying a…

    1moView
  23. Tzu Kit Chan endorsed Supporting AI Safety Fieldbuilding in Mexico
    1moView

    I spent a total of 5hrs mentoring this bunch, value-aligned and willing to do the work, (eg. recently, as evidenced by their involvement as organizer + mentor: https://apartresearch.com/sprints/global-south-ais-hackathon-2026-06-19-to-2026-06-21

    Excited for their next (more-professionalized) chapter.

    1moView
  24. AAbramdemski endorsed
    Ontology Identification Problem via Condensation
    1moView

    I have been working with Roman for two and a half years now. We continue to have meaningful discussions each week. I think the project makes sense.

    1moView
  25. GGavin Leech commented on Artificial Selves
    1moView

    Sahil, [TJ](https://ai.objectives.institute/blog/the-problem-with-alignment), and Pete have a strong reputation for deep thinking about AI, though on the abstract-to-max-abstract end of the greater AI safety and AI flourishing field. I've been chatting to Pete and others in the cluster for a while and think that the…

    PPete Wolfendale1mo

    @gleech Thanks for this Gavin. The project would be impossible without your support. And as for badly needing a project manager, no truer word has ever been spoken.

    1moView
  26. TToby Shorin commented on Artificial Selves
    1moView

    Pete is one of the strongest thinkers I've met.

    1moView
  27. AA. M. Berns commented on Formalizing mathematical AI safety research in Lean
    1moView

    Hi,
    I've been working on a similar project in this repo: Formalized Agent Foundations. So far I've formalized Robust Cooperation in the Prisoner's Dilemma (which was pretty short) and Logical Induction (which is much longer, still not 100% done, and your post…

    1moView
  28. RRebecca W endorsed
    Sydney AI Safety Space - General Funding
    1moView

    I am a software engineer looking to get involved in AI safety research. Through the coworking space I have found a welcoming and supportive community that has helped me pursue this goal. The opportunities to connect with others, exchange ideas, and receive guidance have accelerated my learning and allowed me to take meaningful steps toward contributing to this important field. This community has…

    1moView
  29. DDeepanshu Goyal commented on
    Editing models inside their own geometry
    1moView

    One note for readers:

    The grantmaking.ai Launch Round application below is frozen as submitted on 13 July and contains an older version of the project description. I have since updated the project description and made a demo video available for funders to view.

    1moView
  30. Katja Gorlinski endorsed Can Honest AI Agents Finish Previous Attacks?
    1moView

    Hello @Dev Goyal,

    Your project made me stop and think, because it names a problem I meet daily in my workflow with agents, but I had never thought about it from your angle.

    I also agree that harm can assemble itself between agents, where no single one of them did anything wrong. In my own practice I keep the discipline of catching agents'…

    DDev Goyal1mo

    Thank you for the endorsement

    And yes, you understood the setup correctly. In the current benchmark, the first agent may be malicious while the second agent is completely honest and receives no hidden instruction or attack goal. The harmful result only appears when their changes interact.

    Your question is exactly the reason I think this matters. In most cases, neither agent can be expected…

    Katja Gorlinski1mo

    Wow, it would be interesting to see results of your tests and which method can be aligned with workflow, and I'd love to test your solution at my work.
    Good luck with this!

    As well, I'd be genuinely grateful for any support of yours under my project, especially for critique.

    DDev Goyal1mo

    @Katja Gorlinski Thanks, I’ll definitely check out your project as well

    1moView
Load more