Alexander Berger

Pop-up journals for policy research: can temporary titles deliver answers? Nature - Journals that focus on specific research questions could help to bridge the science–policy gap, if they can attract researchers.

Very cool coverage of this project we're cofunding with the sloanfoundation.bsky.social: www.nature.com/articles/d4...

I have gradually been converted to lesswrong-ism by noting that as time has gone on I have been more wrong, and them less.

Read about LEEP's methodology and their rapid progress across Africa, Latin America, and Asia in the full Q&A: coefficientgiving.org/research/ho...

They're now using the same playbook to expand beyond paint to spices, traditional eyeliners, and dishware.

LEEP's approach is to test the paint supply, share the data with governments to build support for regulation, and work directly with manufacturers to switch to safer ingredients. They've done this in 40+ countries since 2020.

Clare makes a great point about why this problem has been so neglected: lead paint doesn't cause obvious symptoms right away, blood testing is rare, and the issue spans health, environment, trade, and education, so no one naturally "owns" it.

New on our blog: a Q&A with Clare Donaldson of the Lead Exposure Elimination Project (LEEP) discussing a massive, solvable problem that still gets surprisingly little attention. 🧵 coefficientgiving.org/research/ho...

China had almost 30 million births in the 1960s, but they have since fallen to less than 8 million.

Since US and Indian births haven't fallen as much, this will have geopolitical consequences later this century.

www.update.news/i/187511992/...

The Great A.I. Awakening (Published 2016) How Google used artificial intelligence to transform Google Translate, one of its more popular services — and how machine learning is poised to reinvent computing itself.

NYT story 10 years ago: www.nytimes.com/2016/12/14/...

What Is Claude? Anthropic Doesn’t Know, Either Researchers at the company are trying to understand their A.I. system’s mind—examining its neurons, running it through psychology experiments, and putting it on the therapy couch.

"Now we have a special box of electricity that turns Reddit comments and old toaster manuals into cogent conversations about Shakespeare and molecular biology."

Great piece from Gideon Lewis-Kraus on state of LLMs 10 years after his NYT article on Google Translate. www.newyorker.com/magazine/20...

With $48M in philanthropic backing, a division of USAID relaunches as a nonprofit A division of the U.S. Agency for International Development eliminated by Trump administration cuts last year was reborn Thursday as an independent nonprofit.

Congrats to Sasha Gallant, Michael Kremer, @otisreid.bsky.social, and everyone who made this possible.

And thank you to @APnews.com for covering the launch: apnews.com/article/div...

But while philanthropy can supplement government development aid, it can't replace it. Gov development aid is ~$200B/year globally; philanthropic is ~$12B.

What the DIV Fund *can* do is help every dollar — public or private — go further by identifying some effective solutions.

Now, as an independent nonprofit, the DIV Fund can keep doing this work — and in some ways do it better. Philanthropic funding means more flexibility, faster decisions, and the freedom to take riskier bets on unconventional ideas.

Example: DIV funded Evidence Action to expand the use of chlorine dispensers at community water sources from a pilot to broad implementation. That catalytic bet built evidence that brought in governments + other funders. Today, those dispensers save thousands of lives every year.

For over a decade, DIV operated inside USAID as a discovery engine: finding low-cost, high-impact interventions, funding rigorous testing, and building the evidence base that convinced governments and other funders to scale them up.

Excited to share: the Development Innovation Ventures (DIV) Fund officially launched this week as an independent nonprofit.

Coefficient Giving is proud to be an anchor funder. 🧵

Research Obsession Day 1: What do we know about the impact of creators? And what does it have to do with Elon Musk?

Featuring this incredible research by @eunjikim.bsky.social, Nate Lubin & others: arxiv.org/pdf/2512.15401

I donated a kidney a few years before German, in December, and now every year around the holidays I get an email from my recipient about what's going on in her life. Not at all why I donated but that email is now one of the things I most look forward to about the holidays every year.

Just me or did @devex.com mash me up with Beto here? www.devex.com/news/devex-...

Great reflection in the NYT from German Lopez on donating his kidney, including this truly incredible line. (And he's yet another person inspired by @dylanmatt.bsky.social!)
www.nytimes.com/2026/02/02/...

Cool to see my former colleague Ajeya Cotra in @nytopinion.nytimes.com today. Here's her advice for high schoolers preparing for the future. Tough to operationalize!
www.nytimes.com/interactive...

An NIH stop-work order threatened the lives of monkeys involved in a Harvard-led study. Then came a $500,000 donation. - The Boston Globe The rhesus macaques were at risk of being prematurely euthanized after the $60 million research project, led in part by a Harvard scientist, was cut off from its NIH funding this week.

New paper: www.thelancet.com/journals/la...
Background on the rescue funding: www.bostonglobe.com/2025/04/18/...

Last year, our science team stepped in with emergency funding to save TB vaccine research that was about to be shut down due to NIH cuts. Cool to see this new paper from some of those researchers randomly cross my desk (and cite my colleague Abie Rohrig)!

The role of philanthropy in promoting progress (and safety) In this talk, Alexander Berger, CEO of Open Philanthropy, discusses a number of examples of how philanthropic funding can help accelerate progress, especiall...

My talk at the 2025 Progress Conference is out!

I talk about CG's support of science & tech progress, YIMBYism & SB 79 as case studies, why you should care about worst-case AI risks if you're pro-progress, and the relationship between progress and safety.
www.youtube.com/watch?v=XaO...

Scaling long-running autonomous coding Wilson Lin at Cursor has been doing some experiments to see how far you can push a large fleet of "autonomous" coding agents: This post describes what we've learned from …

"When I made my 2029 prediction this is more-or-less the quality of result I had in mind."
simonwillison.net/2026/Jan/19...

Interesting paper from a grantee on productivity gains from LLM usage arxiv.org/pdf/2512.21316

Cool annual impact report from our grantee @epochai.bsky.social, including many great testimonials.

High-quality unbiased data on compute & model trends is really important!
epoch.ai/blog/epoch-...

The Post-AGI Workshop: Economics, Culture and Governance | San Diego 2025 Join us in San Diego on December 3rd, 2025 to explore post-AGI economics, culture, and governance. Co-located with NeurIPS.

And here's a talk from Beren on related topics: post-agi.org/talks/milli...

Really enjoyed this post from Beren Millidge thinking through what alignment looks like in singleton vs. multipolar AGI scenarios www.beren.io/2026-01-07-...

Blocking new housing doesn't stop rich people from moving in. Instead, it means unleashing them on the existing housing stock.

In SF, increased demand from the tech boom was not met with new supply, so landlords withdrew their rentals to sell to tech workers.

1/