Hide table of contents

The AI Safety GiveWiki aggregates donation data from over 200 EA donors and combines them with retroactive evaluations to generate a list of the top AI safety projects to support. Some 50 donors with a total donation budget of $614,401 want to use it to inform their giving.

Please register your donations to improve the platform’s top recommendations, and you may just move some fraction (probably < 10%) of these $614k to a more effective project.

Introduction

A lot has happened since our last milestone announcement. We more than doubled the amount of funding that you, as regrantor/scout, can move; we renamed to AI Safety GiveWiki, which is more descriptive than AI Safety Impact Markets; and we released the above explainer video. (Though not in that order, so please excuse the use of the old name in the video.)


As for the new name, it succinctly conveys what the platform is about:

  1. A crowdsourced recommendation engine for giving.
  2. A place where you can (soon) give away Wiki Credits.
  3. A place to give quickly (wikiwiki ≈ quick) – counter the Donor’s Dilemma.

How to Regrant

There’s no one on GiveWiki who can flick a switch and make you a regrantor. It’s an organic process where you enter your donations to Project X, and if our evaluators are happy with how Project X turned out, then your reputation on the platform will benefit. With greater reputation (donor score) comes a greater weight in our recommendations.

Right now we only have 14 users who have the ambition to inform our recommendations, so it’s still very easy to get started as a regrantor – or “project scout” as we call them.


Concretely, say, you’ve donated to the Center on Long-Term Risk

Find the project page and select the “Register a donation” tab.

Enter the date of your donation, the amount in USD at the time, and optionally the “recommender” – our recommendations, a particular (other) project scout, or your own research?

Click Add. Congrats! You've registered a donation. The charity will be notified of it and will veto it if it's incorrect.

Repeat it a few more times until you've entered all your donations, and you may become a new top donor!


Tips

  1. You have a lot of donations to enter? We can help! Just send us a spreadsheet with the project, date, amount, currency, and recommender to hi@givewiki.org.
  2. A charity you donated to is not yet registered on the platform? Email them and ask them to sign up. It only costs them a few minutes. (But if they’re not in AI safety, the platform might not be the right fit for them at this time.) Failing that, I can create a hidden project for you and register your donations there.
  3. Look for projects with a high “final score.” The higher the final score, the better your donor score. But our evaluators are working on an update to the final scores right now. There will be plenty of changes and plenty of projects that receive nonzero scores for the first time. (Is it weird to call that score “final” if it changes? How about “evaluation score”?)
  4. Like a project a lot? Register your donations to it regardless of its current final score. If your donor score is sizable, you’ll buoy it up in our top ranking!
  5. How much time should you invest? We’re guessing that probably (sadly) only some 10% of our preregistered donors will actually follow through and change their donations based on the GiveWiki recommendations – be it a top project or a project further down that catches their eye. So maybe act like you’re on a committee that allocates $60k. Registering a few years’ worth of donations shouldn’t take more than 30 minutes, so we hope most of you will consider this a good deal.
  6. Regret a donation? Don’t register it. We’ll eventually build out a few features to encourage completeness and allow donors to disavow registered donations. But until then it would be counterproductive to encourage you to enter such donations.

How to Donate

Are you looking for giving recommendations in AI safety? Check out our current top recommendations and:

Please register your interest here!

It’s a super short form that should take you < 1 minute to fill in. Just your name, email, and approximate annual donation budget. Thank you kindly!


Our evaluators will have fresh scores for our projects in time for the giving season, so stay tuned and check back in December to see how our recommendations have changed!

You can also create an account with the platform so you’ll be informed when a new project registers – just like a big funder!

(You’ll need it anyway when you register your donation – which we ask you to do to give us data on how much money we, and the individual project scouts, have moved.)

Epilogue

If you’d like to support GiveWiki and our work on our long-term vision, please get in touch. We’re looking for SAFE-based investments in our public benefit corporation GoodX.

No comments on this post yet.
Be the first to respond.
Curated and popular this week
 ·  · 10m read
 · 
I wrote this to try to explain the key thing going on with AI right now to a broader audience. Feedback welcome. Most people think of AI as a pattern-matching chatbot – good at writing emails, terrible at real thinking. They've missed something huge. In 2024, while many declared AI was reaching a plateau, it was actually entering a new paradigm: learning to reason using reinforcement learning. This approach isn’t limited by data, so could deliver beyond-human capabilities in coding and scientific reasoning within two years. Here's a simple introduction to how it works, and why it's the most important development that most people have missed. The new paradigm: reinforcement learning People sometimes say “chatGPT is just next token prediction on the internet”. But that’s never been quite true. Raw next token prediction produces outputs that are regularly crazy. GPT only became useful with the addition of what’s called “reinforcement learning from human feedback” (RLHF): 1. The model produces outputs 2. Humans rate those outputs for helpfulness 3. The model is adjusted in a way expected to get a higher rating A model that’s under RLHF hasn’t been trained only to predict next tokens, it’s been trained to produce whatever output is most helpful to human raters. Think of the initial large language model (LLM) as containing a foundation of knowledge and concepts. Reinforcement learning is what enables that structure to be turned to a specific end. Now AI companies are using reinforcement learning in a powerful new way – training models to reason step-by-step: 1. Show the model a problem like a math puzzle. 2. Ask it to produce a chain of reasoning to solve the problem (“chain of thought”).[1] 3. If the answer is correct, adjust the model to be more like that (“reinforcement”).[2] 4. Repeat thousands of times. Before 2023 this didn’t seem to work. If each step of reasoning is too unreliable, then the chains quickly go wrong. Without getting close to co
 ·  · 11m read
 · 
My name is Keyvan, and I lead Anima International’s work in France. Our organization went through a major transformation in 2024. I want to share that journey with you. Anima International in France used to be known as Assiettes Végétales (‘Plant-Based Plates’). We focused entirely on introducing and promoting vegetarian and plant-based meals in collective catering. Today, as Anima, our mission is to put an end to the use of cages for laying hens. These changes come after a thorough evaluation of our previous campaign, assessing 94 potential new interventions, making several difficult choices, and navigating emotional struggles. We hope that by sharing our experience, we can help others who find themselves in similar situations. So let me walk you through how the past twelve months have unfolded for us.  The French team Act One: What we did as Assiettes Végétales Since 2018, we worked with the local authorities of cities, counties, regions, and universities across France to develop vegetarian meals in their collective catering services. If you don’t know much about France, this intervention may feel odd to you. But here, the collective catering sector feeds a huge number of people and produces an enormous quantity of meals. Two out of three children, more than seven million in total, eat at a school canteen at least once a week. Overall, more than three billion meals are served each year in collective catering. We knew that by influencing practices in this sector, we could reach a massive number of people. However, this work was not easy. France has a strong culinary heritage deeply rooted in animal-based products. Meat and fish-based meals remain the standard in collective catering and school canteens. It is effectively mandatory to serve a dairy product every day in school canteens. To be a certified chef, you have to complete special training and until recently, such training didn’t include a single vegetarian dish among the essential recipes to master. De
 ·  · 1m read
 · 
 The Life You Can Save, a nonprofit organization dedicated to fighting extreme poverty, and Founders Pledge, a global nonprofit empowering entrepreneurs to do the most good possible with their charitable giving, have announced today the formation of their Rapid Response Fund. In the face of imminent federal funding cuts, the Fund will ensure that some of the world's highest-impact charities and programs can continue to function. Affected organizations include those offering critical interventions, particularly in basic health services, maternal and child health, infectious disease control, mental health, domestic violence, and organized crime.