For a while now I have been recommending anyone working in an adversarial environment to read the book Character Disturbance by Dr. George Simon. Blurb from amazon: psychologist George Simon explains in plain English:
- How most disturbed characters think.
- The habitual behaviors the disturbed use to avoid responsibility and to manipulate, deceive, and exploit others.
- Why victims in relationships with disturbed characters do not get help they need from traditional therapies.
- A straightforward guide to recognizing and understanding all relevant personality types
Smart people I know still regularly fall for personality disordered patterns, and I have seen people I met with personality disturbances successfully apply for grants. I think this is highly underrated and tractable knowledge.
Very excited and much more hopeful about the funding ecosystem being able to focus better on impact rather than PR with this in play. Got it added to the map and funders list, and it'll be in the next funding newsletter.
2The logo for Lightcone Commons is different from the standard Lightcone logo https://www.lightconecommons.com/icon.svg
1Second, I intend to remove requirements for every grant to first receive a speculation grant in order to be considered in a grant round.
In SFF S-Process rounds, Recommenders are allowed to whitelist applications that didn't receive a Speculation Grant, and those applications are considered in the round. The only limits on that (iirc) are that the application has to have been submitted before that round's application deadline, and a Recommender has to evaluate any application that they whitelist.
Oh cool, that appears to be a recent change! The 2025 FAQ indicated more strongly that speculation grants are required. I am glad recommenders can now override this default if they want.
The website does still say:
There are of course always escape hatches, and I am glad they exist, but I wouldn't currently feel comfortable telling anyone that receiving speculation grants is optional in any meaningful way.
I'll change the wording to be a bit weaker though and say "first receive a speculation grant in order to be evaluated in a grant round on the same basis as other applicants".
- No required rubric for evaluating applications
FWIW, I don't know what rubric this is referencing for SFF S-Process rounds.
Broadly speaking, to evaluate a proposal a Recommender needs to make a Marginal Value Function and write some kind of note. We also ask Recommenders to provide some amount of positive or negative feedback to orgs they evaluated.
Oh, I was told this from a recent recommender, but I possibly was wrong! I'll remove it for now and ask for follow-up on what they meant.
Ah I’ve got a guess now: there’s a rubric we added to give AI suggestions of how to improve their notes, with a checklist of 7 features associated with notes that have been most useful in the past.
That’s only one of like 5-10 ways we incorporated AI this year, and the rubric is a nudge rather than a requirement, but I get how someone could have experienced it as “this rubric is the required way to evaluate proposals, yuck.”
Ah, yes, that sounds very similar to what I heard. I reference that in this section a bit, so I think I shouldn't put it back:
Third, I expect to rebuild the tools for getting AI systems to assist with evaluations from the ground up, mostly just because I don’t like the current rubric-based system and want to give evaluators more freedom on how to leverage AI systems in evaluating grants.
Appreciate the clarification!
Furthermore, I want to remove any requirement by evaluators to look comprehensively at all applications submitted to a grant round.
In SFF S-Process rounds, Recommenders are required to skim or familiarize themselves with all applications that are included in the round, but they're certainly not asked to evaluate or look comprehensively at them.
Yep, "look" here was in reference to that required skimming/familiarization. At least historically it also involved every application getting assigned to at least one evaluator to investigate. I agree that not every evaluator is supposed to evaluate every application, which would really be very prohibitive.
fwiw the natural way I parsed this sentence was as claiming that in SFF that each evaluator has to look at all applications, and indeed give them a comprehensive look. I now think you're trying to say that in SFF each evaluator has to look at all applications, which is itself comprehensive, and you're requiring neither that evaluators look at all applications, nor that any given application get a comprehensive treatment
My understanding is:
- Every evaluator is supposed to skim/familiarize themselves with every application
- Every application gets investigated by at least one evaluator (where investigation involves writing at least some notes, and at one point involved giving some positive or negative feedback)
I think this is reasonably well described by "evaluators need to look comprehensively at all applications", but happy about rewordings that are clearer to you.
Sure, I don't personally feel that strongly -- mostly I was trying to intercept a potential miscommunication and registering how I initially parsed it. I think squidging those two claims into one statement is kind of hard. Maybe: "for all applications to get a comprehensive evaluation, or for every evaluator to look at every application"?
1Is it ok to link-post this on the effective altruism forum?
Also curious about why this was marked as "personal blog"
Yep, feel free to link post!
It's marked as personal blog because it's an organizational announcement, which we mostly keep off the frontpage. See Site Guide: Personal Blogposts vs Frontpage Posts.
1
TLDR:
I am launching Lightcone Commons, our software-first platform for distributing philanthropic funding. We connect funders with giving opportunities, coordinate splitting the bill with others who want to fund the same projects, and make it easy to defer to grant evaluators who vet applications and scout for new grantmaking opportunities. Funders can leave and join the platform at any time and without the need to commit any funds in advance.
I and my team (together with SFC, Andrew Critch, and others) have been developing software used to distribute over $100M in grants through SFF and Lightspeed Grants. This new platform brings that technology (and much more) to a broader audience.
Finding high-quality philanthropic opportunities is difficult. Potential grantees are incentivized to pitch themselves aggressively, often making evaluation adversarial. Many of the most effective grants are small. Information about grantees is scattered across the internet, and it is often very hard to tell how much room for funding an organization has. Successfully giving away many millions of dollars without falling prey to grifters, confidence games, or plain bureaucratic conformity, is a great challenge that the vast majority of philanthropists fail.
When faced with these problems, many funders ask family members or other trusted long-time friends to manage their giving. But misalignment between a philanthropist and their foundation is common, and philanthropic foundations rarely manage to attract world-class evaluator talent.
To solve these problems, we solicit thousands of applications from organizations and individuals aiming to do highly cost-effective work across a wide range of worldviews (usually with a focus on benefiting humanity as a whole, on a large scale), and we maintain a network of evaluators with strong track records to investigate giving opportunities (including potential grantees that didn’t apply).
We facilitate debates, arguments, and collaborations between evaluators to improve grant recommendations. Funders can watch and read those (both live and in recorded form), to help them become calibrated about the quality of evaluator recommendations and to help them decide how to delegate their funding decisions (if at all). Any participating funder is welcome to bring their own evaluators into the process.
We plan to organize a funding round every 3 months. As an interested funder you gain full access to all applications and almost all evaluations before you make any funding commitment. Our web platform lets you simulate how you and others would split the bill on any grants multiple people want to fund and where exactly your money would go if you decided to contribute to the grant portfolios of any participating evaluators.
Funder FAQ
When is the next funding round?
Applications for the first round open today! To be eligible for this grant round applications have to come in before August 22nd, and we will announce grant recommendations by October 22nd. From then on we aim for one funding round every 3 months with rolling applications (so the same application still works if you apply after the 22nd of August, you’ll just get a response 3 months later).
How much money do you plan to disburse?
For the first round, I am expecting participating funders to disburse around $15M–25M. Hopefully similar amounts every 3 months (and growing over time).
Who are the funders participating in the first round?
Funders can withdraw and join the round at any time, so this is in-flux, but our current tentative list of funding for the first round is:
How do I sign up to participate in this or the next grant round as a funder?
Please book an onboarding call with me right here. While we place no strict requirements on who can access the platform, I want to make sure we are on the same page about treating applicant information as appropriately sensitive, and that you assign some substantial probability to distributing at least $50,000 in the coming round.
Do you take any fees?
Evaluators receive 2% of any funding directed through them, and we additionally charge a 3% platform fee (for a total of 5% on top of the grant amounts). We lower platform and evaluator fees for funders who distribute large amounts of money through the platform.
How does this whole process work at a technical level?
The funding process works on two simple principles:
There are two main classes of participants in the Lightcone Commons: funders and evaluators.
All funders and evaluators enter a funding order, i.e. the ideal order in which they would distribute money in the round. For instance:
"Project Alpha is most effective, so I want to fund that until marginal money given to it is as good as funding project Beta, at which point I want to fund them both until they are as good as project Gamma, etc. until eventually I run out of money".
Evaluators do so for projects, and funders have the extra option of doing the same for evaluators. For instance:
"Alice seems to have the best judgement, so I want to fund her until her recommendations are only as good as evaluator Bob’s recommendations, at which point I want to fund them both until their recommendations are only as good as Carol’s, etc."
When two funders want to fund the same evaluator, they split the bill in proportion to their total budgets. When two evaluators/funders want to fund the same organization, they too split the bill in proportion to their total budgets. If one funder wants to fund a project for up to $100k, and the other one wants to fund it to $150k, then they split the bill on the first $100k, and the second funder who is more excited about the project and wants to fund it more, covers the remaining $50k.
A virtue of this system is that it tells us immediately how to reallocate funds if anyone's funding order changes, or a funder's total budget changes, while following a simple rule for ensuring that costs get split fairly between participating funders.
If you want to build a more intuitive understanding, you can play around with this widget:
How much time and effort should a funder expect to put into a grant round?
The key thing I recommend funders spend their time on is to figure out which evaluators they trust to produce good grant recommendations, and when the recommendations by an evaluator hit diminishing marginal returns. If you brought your own evaluator to a grant round, then you could make a final funding decision in as little as 2-3 hours of getting yourself familiarized with the process, the UI, plus your onboarding call.
If you do not have a strong pre-established trust relationship with any evaluators in the round, I recommend that you budget on the order of 10 hours to read evaluations and meeting notes or watch recordings of debates and discussions between the evaluators, and to read other content written by the evaluators that helps you understand their worldviews.
You can also participate by funding organizations directly, without deferring to any evaluators. In that case, I recommend more like 50 hours to have enough time to read through applications, ask organizations questions, and make a proper grantmaking portfolio.
But ultimately, if you end up finding that you don't have the time to come to a sufficiently robust assessment of any evaluator or organization to make you think giving to them is a good idea, you can just decide to reduce your giving. You are not committing to any time investment by signing up as a funder.
Who are the evaluators currently providing grantmaking portfolios on the platform?
We actively recruit people with expertise and competence to evaluate grants on our platform. Some evaluators currently confirmed are:
And I am actively working on getting more evaluators on the platform, with examples of who I would love to have on board being:
(Just to avoid any ambiguity, the people in this second list have not agreed to participate as evaluators! They have not endorsed this process! But it seemed useful to give a realistic sampling of people I am aiming to get on board.)
Do you only generate grant recommendations to recognized 501(c)3 organizations?
No. We do not require applicants to have 501(c)3 status, and coordinate grants and investments to for-profits, individuals, educational institutions, 501(c)4 organizations, non-US organizations and individuals. Funders can specify arbitrary restrictions on what kind of organizations or individuals they want to fund, depending on what makes sense from their ethical, legal or tax perspective.
Do evaluators only evaluate grants to organizations that apply via the Lightcone Commons application process?
No. Any evaluator can recommend grants to any organization (or private individual) that they think is a good funding target. We only ask the evaluator to fill out basic information about the organization, such as what kind of legal entity they are, and to perform some basic due diligence to avoid fraud.
We have also imported recent live applications from the Long Term Future Fund, Manifund.org and grantmaking.ai, and we are hoping to share more applications with other funding bodies in the future, to make life easier for applicants.
I run a philanthropic regranting program/foundation/fund. Does it make sense for me to participate in Lightcone Commons?
Yes! Many of the funders participating in our first round are funds or foundations, not just individuals (such as the Long Term Future Fund and the ARM Fund).
You can have your staff join as evaluators, and use your existing decision-making processes to decide your preferred funding order. You can import any existing grant proposals you are evaluating into the round, which means you get to potentially split the bill with other funders on those grants. You also get access to other institution’s funding applications, increasing your pool of potential funding targets, and get to read and learn from the grant evaluation practices of other evaluators and institutions.
What happens at the end of a grant round?
When a grant round gets finalized, every funder is asked to lock in their final amount of funding and allocation order. At that point each funder can see what their final grant recommendations will be. When they are locked in, the funder commits to us and the other participating funders and evaluators that they will send out the funds for these grants in the next 2 months. If they fail to do so, we bar them from ever participating in any Lightcone Commons round again (and will try our best to find replacement funders for the affected organizations).
How do you coordinate multiple funders each wanting to review and change their final grant allocation?
When any funder makes a change to their final budgets and allocations, all other funders get a chance to do the same in response. Only if everyone explicitly approves their final grant allocation do we move forward with the end of the round. If this somehow ends up going in cycles, I will coordinate some kind of compromise, or make an executive decision to remove a funder from the round (though this seems to me very unlikely to occur).
Do I need to facilitate sending out the money myself? That sounds like a lot of work.
No. For charitable donations, we have a nonprofit that can receive your donation and handle sending out the grants, final due diligence, and logistics. For grants to non-501(c)3 entities, we have an LLC that creates and signs contracts with target organizations. The costs for these services are covered by the 3% platform fee on all recommendations (which to be clear is charged regardless of whether you make use of these services).
Both of these are optional and you can also make whatever donations or grants directly from your own DAF, LLC or private bank account (though this does not reduce the platform fee).
Do you ever block grants to any organizations or individuals?
We pose no restrictions on organizations that evaluators want to recommend grants to, and provide no mechanism for evaluators or funders to block grants by other funders. I aim for Lightcone Commons to be a platform that facilitates coordination between funders, evaluators and organizations with a wide variety of worldviews, many of which will strongly disagree with each other. I think the world will be a lot better if we can nevertheless successfully coordinate, critique each other's plans, and split the bill whenever many people want to see the same thing funded, even if they diverge strongly on other issues.
(If you wish to use the nonprofit and LLC grantmaking services, I will do my best to make whatever grants you want to make, but do maintain the option to refuse to make any specific grants, and will occasionally face legal restrictions on the grants we can make)
What if a project needs funding sooner than 3 months from now?
In addition to facilitating funding rounds every 3 months, we also make it easy for funders to run a “direct grantor program” where invited grantors get a special budget that allows them to make grants immediately and unilaterally. This enables more active grantmaking (where the grantors are able to take the time to talk with and get to know a grantee), and allows grantors to use this unilateral grant budget to insure projects against the risk of not getting funding in the next grant round (which enables more risk-taking).
When a grantor funds a project like this, let’s say for $10,000, then in the next grant round the first $10,000 allocated to that project will, instead of going to the grantee, be used to reimburse the grant budget of the grantor.
Beyond that, funders are encouraged to increase the grantor budget of those who made grants that ended up being evaluated as highly impactful, and partially reimburse the budget of those who made grants that almost made the cut of the grant round. We provide UI on our platform to make this easy.
Funders can specify what level of review they want such unilateral grants to undergo before they go out. If they use our nonprofit and LLC grantmaking services we can accommodate both a manual review by the funder, a simple checklist process run by our staff, or a full unilateral system where grants go out as soon as possible.
How do I create a direct grantor program?
You can invite anyone you want as a direct grantor onto the platform in the “Direct Grants” section of the app.
Any grantor who participates in another funder’s direct grants program can also choose to be listed as available for being a grantor for other funders. One of the easiest ways to get started with a direct grantor program is to look through the existing list of direct grantors, and give the ones you think will make good grants by your lights a direct granting budget.
I am a speculation grantor at the Survival and Flourishing Fund (SFF). Can I use my budget to make unilateral grants to Lightcone Commons grantees?
There will probably be some allowance for this; see the Grantor Program FAQ addressed at SFF speculation grantors here.
Are all evaluations shared with everyone?
It is up to the evaluators to decide who to share their evaluations with. The platform encourages evaluators to make their evaluations visible and transparent to all participating funders and other evaluators, which makes it more likely that funders will decide to defer to their grantmaking portfolio, and more likely that other evaluators will engage critically with their evaluations. But we provide tools to make evaluations only visible to subsets of funders and other evaluators, and if some evaluations involve sensitive information, or we grow to have a large enough number of participating funders such that expectations of privacy become an issue, I expect more evaluators to limit their evaluations to a narrower audience.
Are the evaluators compensated for their work?
Yes. Evaluators receive the 2% fee charged on any grants given based on their recommendations.
Sometimes a funder wants to bring in an evaluator who is already on their payroll in some fashion; in that case it’s between the evaluator and the funder to determine what to do with the 2% evaluator fee (which should plausibly go to whatever organization employs the evaluator instead of the evaluator directly, or should just be waived for the case of the funder directing money through their own employed evaluator).
Applicant FAQ
What kinds of projects do you expect to fund?
That’s not up to us! Since any participating funders maintain full discretion about where their funding goes, there is no super short and simple description of what projects will get funding. That said, I am happy to say a few things about what the funders who are currently participating in the platform have historically funded and have publicly said about their giving priorities.
One of the biggest funders in the upcoming round is Jaan Tallinn who has conveniently written publicly about his philanthropic priorities. He has also been the primary funder behind the Survival and Flourishing Fund, which has made all of its grant recommendations over the last 6+ years public.
At a high level, I would summarize Jaan’s giving as “a very intellectually diverse approach to trying to shape the long-term future of humanity”. His giving has historically involved recruiting people he respects with quite different perspectives on the world, and giving them lots of latitude on what they can recommend funding to.
This has produced grants ranging from “advocacy to give legal rights to rivers” to “longevity research” to funding for pronatalist.org, and AI x-risk reduction work at Palisade Research. The bulk of his funding goes to work aimed at aligning, controlling or otherwise shaping superintelligent AI systems, but the tail of his giving is very diverse and it’s hard to rule out much ambitious charity.
One of the other big funders for the first round is Dustin Moskovitz. His most recent essay about his giving is probably most helpful for understanding his perspective. Coefficient Giving (previously known as Open Philanthropy) has also facilitated over a billion dollars of grants of his, many of which are publicly available on their fund pages.
The Long Term Future Fund and the ARM Fund are both contributing ~$2M each for this round. They both focus almost exclusively on how we can shape the development of AI in a positive direction, but are also both open to infrastructure-level interventions to achieve that aim, such as by funding Effective Altruist, Rationality or AI-safety community building, supporting U.S. voting reform, and occasionally the LTFF might also make a grant to avoid other existential risks, like engineered pandemics.
The rest of the funding comes from smaller donors or anonymous donors whose giving I know little about! My guess is they will also skew pretty focused on AI existential risk topics, but I really have a lot of uncertainty here.
Do I need to be a nonprofit to apply?
No! We can make grants to 501c3s, individuals, and for-profit organizations. For individuals, we will determine whether your work qualifies as charitable, and provide you with fiscal sponsorship if so. Some of our funders can only make charitable grants (but others aren't limited to charitable grants, so you will be considered in either case).
Do I need to be in the US to apply?
No! We accept applications from anyone internationally. However, there are some countries where it is in my experience approximately impossible to send money to (in particular Russia and India), so it’s pretty likely we can’t make a grant to you if you are located in one of those places.
What's the deadline for applications?
We accept rolling applications, with a deadline two months in advance of when grants are recommended. The current round closes on August 22, 2026, for grant recommendations made by (roughly) October 22, 2026. Any applications after that will get a response by (roughly) January 22, 2027.
What does a typical application process look like?
We try hard to make the application process quick and straightforward. The application has only four required sections:
You can structure your answers to the first three questions however you want, and the application interface is a live-editable collaborative document.
We might also already have an application from you, if you applied to the Long Term Future Fund, Manifund, or grantmaking.ai in the past few months. If so, you should have received an email with instructions on how to get edit access to your imported applications.
You can update and edit your application any time after you submit it, so don’t worry about getting it all perfect immediately. If your plans have changed, we encourage you to revise your proposal accordingly.
After the application deadline, evaluators and funders using the platform will consider your proposal and make recommendations for how much funding they think it should receive. You can read more about our evaluators and how recommendations work in [section]. Any applications received after the deadline will be evaluated in the next round, three months later.
Most projects will hear back when we announce recommendations, usually two months after the application deadline (October 22nd for the first round). You should receive money within a month after that (and technically you could fail due diligence as we try to send you money, though I expect that to happen very rarely, and it can take longer if you are located outside of the US).
Some projects may hear back shortly after submitting, if your project appeals to a direct grantor (individuals who were given a budget by a funder to unilaterally make grants). Direct grants will be limited for the first round, but our goal is that in future rounds most successful applicants will receive at least partial funding within 1-2 weeks of applying.
Can I submit multiple applications?
You can submit multiple applications for your organization, but please try to avoid them being mutually exclusive or overlapping too much. We might consolidate applications ourselves on the backend if you submit multiple, and leave it up to you how to allocate funding between your projects if you do get funding.
Evaluator FAQ
What's the time commitment for being an evaluator?
You'll likely need at least 40-50 hours of up-front work when you first start evaluating, plus 20–30 hours each round (quarterly) to update your recommendations. This is a lower bound! Many evaluators will spend much more time than that.
The up-front work consists of getting familiar with the applicant pool and writing about your taste and strategy, to help funders decide whether they want to defer to you. After that, you'll have new applications to read each round, but you'll be making updates to an existing set of priorities, not starting fresh.
Will I be compensated for evaluation work?
Yes, evaluators are paid 2% of any funding directed through them.
What does evaluation work involve?
In short: read applications, and rank them in a priority order for funding.
You'll produce three kinds of outputs:
1. Your funding priority order (most important)
Your funding order is the answer to the question “in what priority order would I recommend distributing funding if I was the only funder in this round”.
This kind of ranking allows your priorities to be compared directly to other evaluators', and used to allocate a variable amount of funds. Funders defer to evaluators by placing you in their own funding order, so the amount flowing through your recommendations is not known in advance and shifts as funders join, leave, and adjust their allocations. For the details of our algorithm see “How does this whole process work on a technical level?”.
Your funding order can and should be considerably longer than the amount funders allocate to you, given the way collective funding is allocated. When several evaluators want to recommend the same organization, they split the bill in proportion to the budget they're working with, so your recommendation tends to go further than expected.
2. Explanation of your funding priorities
You are strongly encouraged, but not required, to explain your funding order. Explanations will frequently consist of writeups about specific organizations for other funders and evaluators to read, based on each organization's application. But high-level, general material is also useful – your views on cause prioritization, what you look for as signs of strong teams or projects, and your predictions about the future will all help funders understand the thinking behind your recommendation, and feel more confident in deferring to you.
3. Engagement with other evaluators
You are strongly encouraged to engage with other evaluators – reading and commenting on their explanations, answering questions on your own, and debating points of disagreement. Applications and evaluations live in documents that support top-level and inline comments, with configurable visibility, and the platform includes features for notifications and direct messages to help evaluators manage many parallel threads. We also encourage recording and posting synchronous debates, that other funders and evaluators can watch.
Can evaluators be anonymous or pseudonymous?
Technically, yes. You can sign up with either your real name or a pseudonym, which others on the platform will use to track your priorities and commentary across applications and rounds.
I expect pseudonymity to be most useful when specific funders bring in evaluators who wish to remain anonymous. Currently, the only other way to become an evaluator is for me (Oliver Habryka) to personally select you to participate, which is less likely for evaluators who don't want their identity revealed. Pseudonymous evaluators are also less likely to be given additional funding to allocate by new funders, since it makes it difficult to find other information about them outside the platform.
Who can see an evaluator's presence, funding priorities, explanations, and comments?
By default, your identity on the platform will be visible to all other evaluators and funders – and we aren't asking anyone to commit to secrecy about which evaluators are part of the program.
Your funding priorities, explanations, and discussions have a few different options for visibility. By default, your funding order will be visible to all funders and evaluators on the platform. Other documents you write – including explanations of your reasoning, and evaluations of specific organizations/applications – can have more granular visibility settings. Options for those documents are:
Your comments on other evaluators' work will be visible to everyone who can see the document you commented on – usually everyone on the platform, but sometimes only the specific set of people they've shared it with.
What tools are available for screening and evaluating applications?
All applications are read by a screening team as soon as they come in, so by the time you read an application, it will be tagged with:
In future rounds you'll also be able to see past comments about organizations that have applied, from evaluators who made their comments broadly visible.
We give all evaluators a generous LLM token budget and a set of AI tools we've developed to help evaluate applicants. We don't require you to use a specific rubric for evaluation, so you can build whatever workflow works for you.
You also don't need to read every application. Your funding priorities will be more useful to funders in proportion to how many applications they cover, but you're free to stick to your areas of expertise and ignore applications outside of it.
Can an organization be an evaluator?
Yes! A fund, foundation, or research group can participate as an evaluator, and receive the 2% evaluator compensation as an organization.
I've been an evaluator for SFF. How is the Lightcone Commons process similar and different?
The basic process is the same: evaluators submit funding priorities and comment on other evaluators' priorities, and projects are funded in proportion to recommended budgets.
The biggest change is ongoing engagement: participation in Lightcone Commons is intended as an ongoing engagement, as opposed to the round-based engagement in SFF rounds. Evaluators are welcome to only participate for a single round and then become inactive, but they are also welcome to keep submitting evaluations as an ongoing long-term activity.
The other differences are:
Misc FAQ
What are the biggest changes from the way the Survival and Flourishing Fund has been organizing their funding rounds?
I aim to reach almost complete feature parity with the SFF S-Process, but am planning to substantially redesign a few non-core parts of the process where I am dissatisfied with the current implementation.
First, I want to remove special code and special treatment around matching pledges. Grantees are welcome to indicate in their application that they are seeking matching pledge funding, but evaluators should evaluate whether those matching pledges increase the cost-effectiveness of a grant the same way they evaluate all other evidence about the cost-effectiveness of a grant.
Second, I intend to remove requirements for every grant to first receive a speculation grant in order to be evaluated in a grant round on the same basis as other applicants. Furthermore, I want to remove any requirement for all applications to get a comprehensive evaluation, or for every evaluator to look at every application. Evaluators are encouraged to build grant portfolios that limit themselves to domains where they have expertise.
Third, I expect to rebuild the tools for getting AI systems to assist with evaluations from the ground up, mostly just because I don’t like the current rubric-based system and want to give evaluators more freedom on how to leverage AI systems in evaluating grants.
Fourth, I want to remove requirements for evaluators to attend meetings and engage with other evaluators in highly specific ways. It is OK for an evaluator to be on the platform and never talk to anyone else (though doing so will likely result in funders trusting them less and distributing less funding through them). I want to create affordances and incentives for engaging with the evaluations of other funders both live and asynchronously, but not make it a requirement.
Fifth, I am not planning to have any kind of special tracks, or special grant rounds focused on sub-areas, which feel more opinionated than makes sense for neutral infrastructure like this (though individual funders and congregations of funders are welcome to do special promotions or requests for proposals, and we’ll likely help at least with the basic logistics of those).
Is the code for this going to be open source?
Eventually yes, but I am not guaranteeing that we will have made it open source by the first grant round. A project being open source introduces a few more security risks (such as accidentally leaking credentials or tokens in the code) that I would like to not have to think about early on, but I am aiming to make it possible for someone to fork the code behind this project and run their own copy of it within a few months of getting started.