Tuesday, August 22, 2017

Does AI Safety (and the Effective Altruist Technocracy) Need More of a Grassroots?

The Future of Life Institute released a letter today to the UN’s Conference of the Convention on Certain Conventional Weapons (CCW) conveying concerns regarding lethal autonomous weapons (signed by Elon Musk and covered in The Washington Post and elsewhere). The concerns are grave:
Lethal autonomous weapons threaten to become the third revolution in warfare. Once developed, they will permit armed conflict to be fought at a scale greater than ever, and at timescales faster than humans can comprehend. These can be weapons of terror, weapons that despots and terrorists use against innocent populations, and weapons hacked to behave in undesirable ways. We do not have long to act. Once this Pandora’s box is opened, it will be hard to close. We therefore implore the High Contracting Parties to find a way to protect us all from these dangers.
I used to be unduly dismissive of far future concerns (as did many other EAs), but I've been persuaded by books like Superintelligence that getting artificial general intelligence right is one of the most pressing global problems. Given the danger an AI smarter than humans poses, preventing an AI from having lethal weapons at its disposal seems like a really, really big deal.

If that's the case, then I have a knot in my stomach about the circumstances surrounding the UN's decision-making on the matter. A ban on "killer robots" (not explicitly called for in the letter but something the WaPo and others took as implied) is not an easy policy for a government to stomach. What happens when the UN rejects the proposal and adopts an overly weak one that leaves the world as unprepared for killer AI as it was for nuclear weapons?

With many of the countries signing the convention being at least partially democratic, I wonder if public pressure is part of the answer. Even for non-democratic governments, public pressure can matter. I've previously made the case, which I believe still stands for pressing causes, that collective action is an effective way of creating change. Is there a need for a grassroots movement on this issue?

I know many people who have studied this more than me disagree, but there needs to be a way to translate expert opinion and knowledge into policy. How do we do this, and why is or is not a public movement part of the answer?

Monday, August 21, 2017

The Washington Post's Omission on Voter Suppression

The Washington Post this weekend had a sharply-worded editorial condemning efforts to disenfranchise voters with which I of course agree, but the editorial missed the crucial point. Here's the key paragraph:

Yet even if all 1,500 Confederate symbols across the country were removed overnight by some sudden supernatural force, the pernicious crusade to roll back voting rights would continue apace, with voters of color suffering its effects disproportionately. Pushing back hard against those who would purge voter rolls, demand forms of voter ID that many Americans don’t possess, and limit times and venues for voting — this should be a paramount cause for the Trump era.
All the forms of voter suppression listed deserve to be combatted, but the chief cause of disenfranchisement in the United States is one that is still rarely challenged: disenfranchisement of those with felony convictions. It's difficult to get a good estimate of the number of people prevented from voting by the most-talked-about forms of voter suppression, but estimates put the number in the hundreds of thousands or low millions. Felony disenfranchisement alone revokes the right to vote for 6.1 million people according to the Sentencing Project, including nearly 10% of African Americans. Of course, many if not most of those barred from voting would not vote anyway, but if even a third did, that would likely outweigh the number of people disenfranchised by all voter ID laws, voter roll purges, and other restrictions.

Felony disenfranchisement, in turn, has huge knock-on effects by enabling the continued incarceration of a massive portion of the American populace and a particularly large portion of the African American populace. Any effort to fight for voting rights in the 21st century that wants to do more than tinker around the edges needs to fight felony disenfranchisement.

Friday, August 18, 2017

How Should Animal Advocates Think About Anti-Big Ag Political Coalitions?

Source: http://www.nyanimalag.org/wp-content/uploads/2013/01/CowsFeeding5.jpg
Pooh-poohing the weakness of mainstream political parties toward Big Ag is rightly a hobby for animal advocates. The laxness (or supportiveness) of U.S. federal and state agencies toward Big Ag is strikingly out of step with the American publicAnimal advocates should be heartened this week by a piece in The Nation that argues that Democrats can seize the current anti-corporate fervor to challenge the Cargills and Smithfields of the world. The wrinkles of the piece, though, lay out some more complex strategic challenges than animal advocates may notice.

The article introduces us to Joe Maxwell, "a fourth-generation hog farmer and former lieutenant governor of Missouri" who led the fight against Oklahoma's "Right to Farm" bill, which ended in a rout of Big Ag. A "Right to Farm" bill is well worth fighting. So too are efforts to up the ante on USDA regulations and enforcement.

Then the piece gets to the biggest goal:
But the biggest demand from Family Farm Action is for the government to reinvigorate the antitrust laws that ensure open competition and prevent collusion. A major case involved the rigging of a key benchmark price grocery stores use to buy poultry, which cost consumers hundreds of millions of dollars. “If the legal definition of collusion doesn’t give the Department of Justice the ability to prosecute, then we need to change the laws,” Maxwell said. The organization also supports using the Sherman and Clayton Acts to break up concentrated agricultural markets.
The catch here should be evident to many: anti-competitive practices, collusion, and price gouging may generally sound like bad things, but one key effect of them is to decrease the amount of a "product" being produced: that is, they should lead to fewer animals being raised. For consumers, small farmers, and potentially environmentalists, anti-competitive practices are indeed bad, but for animal advocates it may be a good thing that agriculture is so monopolized.

That does not necessarily mean that animal advocates should hop off this political bandwagon. It may be worth trading this issue for others. Even on this demand, though, there are more wrinkles than the number of animals raised. Monopoly power gives the industry extra money with which to lobby politicians, push anti-animal policies (even laws penalizing animal advocates), and get away with more egregious abuses. That said, it's not clear how much government policies actually do prop up industry. The Open Philanthropy Project's Lewis Bollard looked into a few of these policies and found that they did not make as much of a difference as many of us think.

There may be a greater reason for a political coalition, though, that outweighs the concern about how such a coalition would affect the numbers of animals raised for food in the short term. An anti-Big Ag rural coalition may reshape the social landscape in ways that down the road may lead to greater change. Connecting animal advocates with greater resources and voters could pave the way for bigger battles ahead. Public mobilization around Big Ag may be far more critical to the future of nonhuman animals than the policies of today. If political coalitions can pave the way for public mobilization, they may be worth doing despite problems like those above.

Wednesday, July 12, 2017

What Should We Make of India's Cow Protection Movement?

"[O]n April 21, 2017, in the Jammu region of Jammu and Kashmir state, a mob brutally attacked five members of a nomad cattle-herding family, including a 9-year-old girl, on suspicion that they were taking their cows for slaughter. A video posted on social media showed a group of men chanting slogans commonly used by BJP supporters, breaking down the family’s shelter, beating an elderly man with rods and sticks even as women begged for mercy, and finally setting the shelter on fire."

This episode is the latest in a trend sweeping India of "cow protection" that has won praise and accolades from some animal advocates, from up and down my newsfeed, and from Leonardo DiCaprio. The tide has swept high enough that India recently banned the sale of cows for slaughter. It's now reached the point at which the movement turns to violence, high on its own success.

What should animal rights activists make of this movement? The turn to violence is something to oppose given studies on 60s urban riots, French labor movements, and civil resistance more broadly. The campaign also deserves condemnation as an instance of Hindu nationalism and anti-Muslim bigotry (not all that different from the alliance between some European animal advocates and anti-Muslim forces).

What should we make, though, of the campaign's impact on animals, as well as campaigns to, say, tax cow flesh because of its impact on climate change?

Many animal advocates are inclined to praise such achievements as incremental changes that lead down the path to protection for all animals. I think this view is flawed. While I'm inclined to look at the path of history as bending toward justice, that does not mean that any policy change necessarily does so. To know whether a policy change will have the right effect, we should think of the way it affects political forces at play in shaping policy.

In this sense, "cow protection" has a few effects relevant to animals:

1) It satisfies the forces pushing for cow protection or a cow flesh ban. In India's case, this is primarily Hindus, although American environmentalists have provided support as well. By satisfying these forces, those forces will be less willing to advocate for a broader policy that includes the goal of banning cow flesh in the future.

(In India's case that may not be this concerning, as an alliance with Islamophobic nationalists is probably not an appropriate move for animal advocates, but with environmentalists this should be a serious concern.)

2) It sets a cultural signal - although this could go both ways. It says that banning animal products acceptable, which can make other bans more plausible, but it also says that cows are a particularly bad animal to exploit, which actually goes against what we should do based on the sheer numbers of animals exploited.

3) It leads to people eating fewer or no cows and eating chickens, fishes, and pigs instead. Given that a "serving" of chicken, fish, or pig flesh contains far more suffering than a serving of cow flesh, this means more suffering and death by more animals. Perhaps because of environmentalist arguments against eating cows, American cow consumption seems far more troubled than American chicken consumption.

By my assessment, the longer-term picture is bleak, although slightly ambiguous, and the short-term picture is less ambiguously bad. Given that, animal advocates should not support the cow protection movement.

Tuesday, March 21, 2017

Trump's Supreme Court Nominee May Be the Most Important Thing Trump Does to Animals

Trump Supreme Court nominee Neil Gorsuch on a fishing expedition with Antonin Scalia
As much as Trump promises to empower the most extreme voices on civil rights, LGBTQ rights, and immigration, animal advocates have long had good reason to fear the Trump administration. Trump's ties to notorious ag empresarios and persecutors of activists from Forrest Lucas of Protect the Harvest to Bruce Rastetter of "ag gag" fame give cause for worry. Despite this all, potentially the most important - that is, damaging - nomination for animals is Trump's Supreme Court pick, Neil Gorsuch.

Why the Supreme Court? The Supreme Court touches on all sorts of issues - LGBTQ+ and women's rights, voting rights, civil liberties, federalism, etc. - but when has it ever played a major role in animal rights? How could we even predict how a Supreme Court nominee would rule?

Trump with nominee Neil Gorsuch
The thing to note is that while nominations, personal ties, regulations, and even laws come and go, it takes a much longer time for a Supreme Court nominee to come and go. At 49, Gorsuch is the youngest nominee to the court in 25 years. He will likely be on the court for 30 years at least - and possibly up to 40 given his health and excellent healthcare. The question is not what cases are before the court today, but what cases will be before the court in the next 30 - or 40 years.

Based on the answers I've heard from leaders of our own movement, most of us think we will achieve something close to animal liberation, or at least the end of factory farming, in that timeframe. This is not just DxE; at a panel at last year's Effective Altruism Global x in Boston, Bruce Friedrich of the Good Food Institute, Jon Camp of The Humane League, and Sharon Nunez Gough of Animal Equality all offered a similar timeline. Add to this that there has not been a major movement for the rights of an oppressed class - not civil rights, women's rights, children's rights, disability rights, or LGBTQ+ rights - that avoided the court, and it becomes exceedingly likely that our movement will have a vital, life-and-death (or rather life-and-mass murder) case under Justice Gorsuch and anyone else nominated any time soon.

What sorts of cases might come before the court? There are an array of possibilities, and there likely will be an array of cases, but here are two:

A screenshot from "Unlocking the Cage"
-Cases regarding the legal standing of animals and rights owed to animals under pre-existing laws. The Nonhuman Rights Project, should it succeed in getting lower state courts to grant nonhuman animals limited rights and legal personhood, will likely see repeated appeals and could very well make it to the U.S. Supreme Court. Indeed, that seems like a necessary step to get the sort of sweeping precedent NhRP would like. Down the road, one could imagine ideas previously seen as ridiculous - like PETA's case against orca slavery under the 13th amendment or a case for equal protection under the 14th amendment - come before the court. Courts are often critical to the protection of groups with less political power from a tyrannical majority. Who has less political power than nonhuman animals?

-Cases regarding the degree to which regulations at both the federal and state level can stand. The Supreme Court has a lousy track record here, having struck down a California animal welfare law, saying that it was pre-empted by a weak federal law, the Federal Meat Inspection Act. In a unanimous ruling, conservatives achieved their anti-regulatory goals and liberals achieved their goal of strengthening the federal government all at the expense of animals, which neither side currently cares about. Hopefully the court's liberals will start to care about animals more as time goes on, but the more conservative the court is to start, the more difficult the climb will be.

There are other types of cases we likely can't imagine that may surprise us, and the best way to prepare for unforeseen cases is to have the best people on the court possible.

Given that we know next to nothing about Gorsuch's views specific to animals and basically never know anything about a nominee's specific views on the matter before they are confirmed, how are we to judge?

Well, despite all the talk in the animal rights movement about how animal rights and other progressive views do not imply each other and vice versa, it's generally true that the more progressive someone is, the more supportive they'll be of animal rights. Groups like HSUS argue that animal rights is a nonpartisan issue while others point out that many animal rights activists are xenophobic, Islamophobic, racist, or sexist.

The truth is that while these both may be true, progressives and people who identify as Democrats tend to be more supportive of animal rights. Animal rights shares an intellectual kinship with progressive values, particularly the sorts of progressive values held most deeply by those trained in reason and logic, i.e. lawyers and judges. So while liberals on the court are failing animals, there's a lot more hope that they'll come around. Another conservative on the court, on the other hand, is a very bad thing for animals - for decades and decades.

Update: Another point worth noting is that Gorsuch has critiqued Peter Singer's utilitarian philosophy at length on the topic of euthanasia. This is yet more reason to think that Gorsuch will be unsympathetic to animal advocates. Thanks to Harrison Nathan for pointing this out.
Want to take action? Call 202-224-3121 and ask to be put through to the Senator for your state. Tell them to filibuster the nomination of Neil Gorsuch.

Tuesday, March 14, 2017

Vox's Piece Against Bill Gates's Chicken Donation Misses A Major Externality

Economist Chris Blattman (a Principal Investigator where I work, Innovations for Poverty Action) writes an on-point criticism of Bill Gates's push for donating chickens at Vox, but he misses a major cost. It's a cost that many of the forefathers of modern welfare economics, like Jeremy Bentham and John Stewart Mill noted well ahead of their time. The cost is the lifelong suffering and complete obliteration of individuals - chickens - who possess all the morally relevant characteristics of economic agents for welfare economists to consider them.

There are many reasons why economists ignore animal suffering when analyzing agriculture. One is likely simple speciesism. There's a deeper set of reasons that I think make economists ignore animal suffering, though, and they show how far modern economics has strayed from its roots in many ways. The reasons start with this: a refusal to make a "moral" judgment.

The basic premise of modern welfare economics is that there is no way to compare how well-off two people are (what's called an interpersonal comparison of utility). This belief stems from mid-20th century economist Lionel Robbins. Of course, the belief is absurd on its face: the idea that there is no way to compare the happiness of a poor Nigerian family with an American billionaire is ridiculous. Economists have leaped from the idea that there is no way to prove or know that the billionaire is better off to the idea that there is no way to have high confidence.

In practice, few economists believe what is taught as a premise in Econ 101. Professor Blattman and just about every economist I've worked with at IPA is in that field precisely because of a belief that people in poor countries are worse off than most people in wealthy countries. Just about every measure of overall economic well-being, like GDP or the poverty rate, lumps an imperfect measure of people's well-being together and assumes that well-being is in fact comparable.

The residue of the idea that you can't compare well-being, though, sticks around, and the complete exclusion of animals from economic calculations is its most radical implication. From the perspective of an economist, a dog or a chimpanzee is no different from a chair. This flies in the face of scientific consensus. The scientific consensus is hardly revolutionary: from Greco-Roman to Judeo-Christian to Buddhist to Hindu thought, the view that animals deserve moral consideration is ancient. The scientists merely confirmed what we all could have presumed.

Interestingly, the reasons to care about animals in economics extend even further: many animals have the characteristics of an economic agent. There's an entire field of animal welfare science that, while it is still rife with bias and support for hurting animals, takes as its basic assumption that animals express preferences, an assumption that economists know as the weak axiom of revealed preference in the human case. Neurological studies of what goes on in animals' brains when they make these choices suggest that this is accurate: where we have measured it, animals' reward and decision-making systems parallel humans' brains when making decisions.

In light of these findings, it is indefensible to classify animals in the same category as furniture in economic analysis. If we do that, then the costs of a program like Bill Gates' stack up immediately. Chickens are generally killed as babies at a fraction of their age. In poor countries like those where he will give these chickens, animals are often slowly hacked to pieces when it comes time for slaughter. Before their death, as anyone who's walked the streets in a developing country will know, birds are confined in wire cages little better than American battery cages (something Bill Gates himself is working to combat).

If that's not enough, though, think about this: at a time when industrialized countries are realizing the toll of factory farming, the giant animal suffering factories of America and Europe are desperately seeking to expand in developing countries. What better way to spread the idea of mass animal product consumption than teaching people to raise chickens? Gates is condemning not just the chickens he distributes but generations of chickens beyond them to lives of misery.

All this suffering is not worth $5 per bird, and that's the optimistic estimate of a chicken "donation" program's effectiveness.
We can quibble about the human costs and benefits of a program like Gates', but the animal cost is glaring.

Professor Blattman writes of some imaginary Gates advisor, "They sold you on the benefits, and didn't tell you how much it all costs." When it comes to animals, the same could be true for anyone who's studied economics and agriculture. When it comes to animal agriculture, economists have been sold on the human benefits but neglect the animal costs.

Friday, January 6, 2017

2016 Was a Good Year and a Bad Year for Humanity, Depending on How You Count It

Though the end of 2016 was greeted by most people I know with a sigh of relief and by pundits as being the end of a very bad year, people who look carefully at the evidence on social trends have been pushing back. Economist Max Roser and representatives of Innovations for Poverty Action (where I work part-time) both recently wrote columns in the Washington Post about why 2016 was, in fact, a great year for humanity. Worldwide poverty continued its massive decline, and there was no great increase in violence despite what people seem to think, leaving us still far ahead of humanity and pretty much any time in the past when it comes to the likelihood of dying a violent death, as psychologist Steven Pinker chronicles in the Better Angels of Our Nature and a more recent interview. All in all, humanity is likely doing better now than we were a year ago, in a continuation of an ongoing trend.

Odd, though, that so many people think things are so bad. At the end of 2015 there was a similar debate following the rise of ISIS (which has continued), the Ebola outbreak (which has largely been tackled), and so on. In 2015, I was inclined to agree with the view that the year had actually been a good one without reservations, but in 2016 I can't. What seems to be going on in many of these conversations is that the public looks at how much violence they see, and the public says, "too much." Then, academics look at how the amount of violence is changing and say, "but it's getting so much less bad!" The unanswered question is whether that change is likely to continue or revert: are things continuing to get better at the same rate, and does it look likely that this will continue to happen, or not? On that metric, I think 2016 was probably a very bad year.
Was 2016 like this...?
...Or like this?  
The rise of nationalist populism in the West, including Brexit, Trump, Italy's referendum, and more, is a marked change, as far as I can see, from the past several decades at least. Not only is it a marked change, though - it's a change from basically all the good trends that have led to a striking decline in violence over the course of human history (again, see Steven Pinker). It's a shift away from international cooperation toward narrow tribalism, away from norms against violence to norms increasingly accepting of it, away from feminism and ideals of equality toward hyper-masculinity and seductive hierarchy, and a shift away from any patina of economic reasoning toward instinct-driven policy.

Now, there are optimistic signs outside of rich Western countries - from Estonia to Ghana - but there are also countries like Brazil and the Philippines whose leaders make Trump seem like a wise statesman.

Worse than anything, though, is that the current trends seem to increase the probability of black swans - rare, catastrophic events with the ability to destroy or inflict misery on vast numbers of people and that books like The Better Angels of Our Nature probably don't quite account for. Trump seems determined to make nuclear war more likely and put exactly the right people around him to make it happen. His team's indifference to climate change in the face of overwhelming evidence does not bode well for other technical risks like an extremely harmful artificial intelligence or nano- or biotechnology run amok.

A few bad leaders is well below the norm for world history, but the increase in leaders bad in such particularly meaningful ways is cause for concern. 2016 may have been a good year for humanity in terms of sheer improvements, but it also brought signs that we might get a lot fewer improvements - and maybe some catastrophic steps backward - in the years ahead. To me, that's what's most striking - and terrifying - about 2016.