These monthly posts originated as the "Updates" section of the EA Newsletter.
You can also see last month's updates, or a repository of past newsletters (including past organization updates).
In December, 80,000 Hours added over 170 roles to their job board and released five pieces of content, including their most popular podcast episode to date, which features David Chalmers:
- David Chalmers on the nature and ethics of consciousness
- Anonymous answers: What bad habits do you see among people trying to improve the world?
- Anonymous answers: What mistakes do people most often make when deciding what work to do?
- Peter Singer on being provocative, EA, how his moral views have changed, and rescuing children drowning in ponds
Finally, they updated their guide to choosing a charity for 2020.
Animal Charity Evaluators
Animal Charity Evaluators recently published updates from organizations that received grants from the Effective Animal Advocacy Fund. They also shared a blog post in which some ACE staff members discuss the reasoning behind their 2019 donations.
Center for Human-Compatible AI
Adam Gleave, Michael Dennis, Neel Kant, Cody Wild, Sergey Levine, and Stuart Russell had a new paper (“Adversarial Policies: Attacking Deep Reinforcement Learning”) accepted by ICLR.
Centre for Effective Altruism
- Applications for EA Global: San Francisco (20-22 March) close on Friday, 31 January!
- CEA’s donor lottery will close on Friday, 17 January!
- CEA published an update on Effective Altruism Funds and a post on how they currently think about EA Global admissions.
CE published the Political Change — Approach Report. The report examines two ideas for reaching out to governments to change laws that affect animal welfare (dissolved oxygen for fish in Taiwan and feed fortification for egg-laying hens in India). Patrick Stadler, CE's Charity Mentor and Curriculum Developer, published a new article: “Six Ways Your Charity Startup Might Fail — and How to Prevent That.”
Effective Altruism Foundation and Foundational Research Institute
In 2016, EAF launched a ballot initiative to drastically increase Zurich’s development cooperation budget. They later coordinated a counterproposal with the city council that preserved essential features of their initiative. This counterproposal passed at the ballot with a 70% majority. The development cooperation budget will thus increase from $3 million to $8 million per year and will have to be allocated “based on the available scientific research on effectiveness and cost-effectiveness.”
- EAF published a research agenda on the Alignment Forum: “Cooperation, Conflict, and Transformative Artificial Intelligence.” It outlines what they think are the most promising avenues for avoiding conflict involving advanced AI systems. They also published a comparison of the state of AI research and development between the U.S., China, and Europe.
- EAF also announced their plans for 2020, including investigating the questions listed in their research agenda, making grants, and hosting research workshops.
Faunalytics has multiple studies in progress and has just begun a new study on reducing the suffering of farmed chicken and fish. They will investigate the frequency and implications of negative beliefs held by the general population with respect to small-bodied animals, identifying the biggest barriers for institutional and individual campaigns. They are accepting donations for all of their current research in progress.
They also published a new blog highlighting their most popular research library entries of 2019. High-traffic entries focused on social issues related to veganism and veg advocacy, TNVR (Trap, Neuter, Vaccinate, Return) programs, and wildlife trauma.
Future of Humanity Institute
Two papers by FHI researchers have been accepted to the Association for the Advancement of Artificial Intelligence (AAAI) workshop on AI safety in February 2020:
- “(When) Is Truth-telling Favored in AI Debate?” by Vojtĕch Kovařk and Ryan Carey
- “The Incentives that Shape Behaviour” by Ryan Carey, Eric Langlois, Tom Everitt and Shane Legg
A paper by Michael Cohen, Badri Vellambi and Marcus Hutter, “Asymptotically Unambitious Artificial General Intelligence,” has also been accepted to the main AAAI conference.
Ben Garfinkel and Allan Dafoe published an article in War on the Rocks based on their paper “How Does the Offense-Defense Balance Scale?”
GiveWell decided to allocate $2.6 million received in discretionary funding from the third quarter of 2019 to Helen Keller International’s vitamin A supplementation program. They also recommended a $355,000 grant from the Effective Altruism Global Health and Development Fund to One for the World, and recommended a $214,425 exit grant from Open Philanthropy to Charity Science Health.
Global Catastrophic Risk Institute
GCRI has announced a new call for advisees and collaborators for select AI projects. They welcome inquiries from students and professionals at all levels who are interested in seeking their advice and/or collaborating with them on AI projects funded by a recent donation for AI research and outreach.
GCRI also recently hosted a session on global catastrophic risk at the 2019 Annual Meeting of the Society for Risk Analysis.
Open Philanthropy Project
Open Philanthropy announced grants including $9.1M to Cincinnati Children’s Hospital Medical Center, $4.2M to the Texas Organizing Project, $1.7M to Anima International, and $705K to the Berkeley Existential Risk Initiative. They also announced their allocation of $54.6M to GiveWell top charities and published a set of suggestions from Open Philanthropy staff for individual donors.
Ought published a technical progress report (Evaluating Arguments One Step at a Time) describing its recent factored evaluation experiments. Ought found that factored evaluation can indeed distinguish some valid from invalid arguments for claims about movie reviews. But a lot of variance exists in the data, which makes the results sensitive to ensembling methods.
Ought is excited about the foundation this work establishes: the methodology and success criteria defined over the course of this experiment should set clear frameworks and baselines for future work.
Raising for Effective Giving
REG’s DoubleUpDrive raised a total of $3,421,524 from professional poker players and other donors, including $640,000 across 80,000 Hours, the Centre for Effective Altruism, the Forethought Foundation, and the Long-Term Future Fund. They will publish an analysis of the Drive’s counterfactual impact later this year.
Rethink Priorities, a project of Rethink Charity, published an intervention profile on ballot initiatives, and continued to publish their analysis of the 2019 EA Survey with posts on careers and skills and cause prioritization among EAs.
The Good Food Institute
- The district judge in GFI's Arkansas label censorship lawsuit granted the organization's preliminary injunction, meaning that Arkansas cannot enforce its label censorship law against Tofurky while the primary lawsuit is pending. The judge held that GFI, the American Civil Liberties Union (ACLU), the Animal Legal Defense Fund (ALDF), and Tofurky are “likely to prevail on the merits of [our] First Amendment claim.”
- GFI submitted a public comment to the National Oceanic and Atmospheric Administration (NOAA) on one of the agency's four new blue technology strategies.
- GFI supported the development of a new XPRIZE competition aimed at producing full-sized, whole-muscle chicken and seafood alternatives. The prize competition will be supported by $13M of funding from the XPRIZE Foundation and is expected to incentivize the founding of more than 10 new companies in the alternative proteins sector by 2022.
- GFI Brazil organized a series of briefings with scientists from the Brazilian agricultural research agency Empresa Brasileira de Pesquisa Agropecuária (Embrapa). This relationship is already paying dividends in scientific interest in Brazil, as GFI has received more Competitive Research Grant Program proposals from Brazil than from any other country besides the U.S.
The Life You Can Save
Since the relaunch of The Life You Can Save, The Life You Can Save has seen 10,000 new email subscriptions and over 15,000 book downloads. They are developing partnerships with a variety of organizations, and encourage readers to get in touch if they would like to help share the book.
The Giving Games Project is currently undertaking an impact analysis based on 2019-2020 participant results, and will make those results available shortly.
Wild Animal Initiative
Wild Animal Initiative had a paper accepted for publication by the flagship journal in the field of restoration ecology; wrapped up Season 1 of the Wildness podcast; and released new research on insecticides, long-term considerations for intervention design, and optimal population density.
At EA Global: London 2019, WAI Research Fellow Dr. Will Bradshaw described how biomarkers of aging could serve as an objective and generalizable metric of cumulative welfare. The recording is now available for viewing.
Add your own update
If your organization isn't represented on this list, you're welcome to provide an update in a comment.
You can also email me if you'd like to be included on the list of organizations I ask for updates each month; I can then add any updates you submit to future posts. (I may not accept all such requests; whether I include an org depends on its size, age, focus, track record, etc.)
(Nitpick: It should say "Foundational Research Institute" rather than "Foundational Research Initiative".)
Thanks! Organizations submit their own names, so I hadn't realized this was a mistake, but I'm glad to have the proper title.
It was written correctly in the Google Doc though ;)
The Google Doc changes every month as orgs update over their old updates. I just copy-and-paste when it's done. ¯\_('')_/¯
Hopefully, the heading just remains unchanged from now on!