Cross posted from the EA Newsletter and effectivealtruism.org

People in the effective altruism community have collectively spent tens of thousands of hours thinking and writing about how to do the most good. In fact, in December 2016 alone, the community produced 50,000 words of analysis on where to give.

I sifted through all the content I could find to come up with the five posts that I thought were the must-read posts from 2016.

This list is based on a synthesis of:

• This post on the EA forum and the nominations that resulted;

• EA Forum’s top posts during 2016,

• Articles featured in past issues of the EA Newsletter;

• articles posted to the blogs of major EA organizations;

• and my personal opinion.

Which articles were ultimately selected is mostly a reflection of my personal opinion and are weighted towards articles that are high on substance. Given the volume of content produced during the year, I likely missed some really good articles that didn’t get posted to a major EA community outlet. If so, please share suggestions in the comments below

Without further ado, here are the top 5 posts in no particular order:

Donor Lotteries by Carl Shulman

Giving money to effective charities is a key part of EA and so it's no wonder that in 2016 the EA community continued to produce lots of content on both where and how to give. Some of the standouts include Rob Wiblin’s end-of-year guide to donating, Owen Cotton-Barratt’s post on using donation commitments to give more now, and any of the posts on where those working in the community plan to give.

But, for my money, the best donation post of the year is Carl Shulman’s post on donation lotteries. The post provides a workable solution for the high fixed costs that small individual donors encounter in determining where to give. Namely, such donors should pool their resources into a donation lottery which allows the winning donor to determine where the entire pool goes. This incentivizes the winning donor to incur the necessary fixed costs to donate the pool effectively while incurring minimal coordination costs.

Also check out his previous post on the subject and his follow-up post on why small donors should be able to use donation lotteries to do at least as well as very large donors.

How Much is One Vote Worth? by Rob Wiblin

The ethical consequences of political decisions became an increasingly popular topic last year, given both the heated US presidential election and the surprise Brexit vote. Some of the content focused on the merits of particular candidates, other content focused on interesting ideas like vote trading or how the results should influence your worldview.

However, my vote for the best content on voting goes to Rob Wiblin’s post on the value of a single vote. The post does an excellent job of just doing the math to show that voting is tremendously valuable if you care about the effect of the president on global wellbeing.

Three Key Issues I’ve Changed My Mind About by Holden Karnofsky

A key virtue in EA is the willingness to follow the evidence wherever it leads. To encourage following the evidence, the community has developed good norms around changing your mind and being willing to say how your mind has changed publically.

Some great posts in this vein from past years include Charity Science’s writeup on grant writing and GiveWell’s mistakes page. But my pick for the most interesting and consequential post from this year is “Three Issues I’ve Changed My Mind About” by Holden Karnofsky (co-founder of GiveWell and the Open Philanthropy Project). In the post Holden shows how his thinking on Artificial Intelligence safety, the EA community itself and what great interventions look like has evolved over time. The post highlights that even if you spend nearly all of your time thinking about how to do the most good, you should still expect your thinking to develop substantially over time.

Deworming Might Have Huge Impact, but Might Have Close to Zero Impact by Sean Conley

One big story in global poverty circles last year were the so-called “Worm Wars,” a debate amongst global health academics about the effectiveness of deworming interventions. It included two well-respected organizations, Cochrane and the Campbell collaboration, saying they find little reliable evidence in favor of mass deworming.

The EA community has been interested in deworming since Giving What We Can recommended SCI in November of 2009, and GiveWell recommends SCI and Deworm the World as top recommended charities, so the Worm Wars were highly relevant to the EA community. Several posts provided an excellent analysis of the topic including “Why I Mostly Believe in Worms” and Giving What We Can’s intervention report on parasitic worms. But I think none states the situation as clearly as “Deworming Might Have Huge Impact, but Might Have Close to Zero Impact.”

Given GiveWell’s reputation as recommending interventions with extremely robust evidence behind them, it may seem surprising that GiveWell think that there is a significant chance that deworming has little impact. In this post Sean outlines the mixed evidence for deworming, but also why deworming is an excellent intervention even given the chance that it may be ineffective.

Concrete Problems in AI Safety by Dario Amodei, Chris Olah et al.

The EA community continued to be interested in AI Safety and continued to produce useful research on the topic. Open Philanthropy Project announced that they were planning to make AI Safety a major focus in the future, the Machine Intelligence Research Institute published an interesting result on Logical Induction and the top-voted post on the EA forum was a lengthy review of the AI Safety literature and charities

However, my pick for the best post on AI Safety this year goes “Concrete Problems in AI Safety” by Dario Amodei, Chris Olah, Jacob Steinhardt, Paul Christiano, John Schulman, Dan Mané. While the post is highly technical, it provides a useful list of practical problems that the field should aim to solve.