Alex319

Posts

Sorted by New

Comments

Incentivizing forecasting via social media

For what it’s worth, while Facebook’s Forecast was met with some amount of skepticism, I wouldn’t say it was “dismissed” out of hand.

 

To clarify, when I made the comment about it being "dismissed", I wasn't thinking so much about media coverage as I was about individual Facebook users seeing prediction app suggestions in their feed  I was thinking that there are already a lot of unscientific and clickbait-y quizzes and games that get posted to Facebook, and was concerned that users might lump this in with those if it is presented in a similar way.

 

Yeah, they certainly would be reluctant to do that. But given that they already do fact-checking, it doesn’t seem impossible. 

I agree, and I definitely admit that the existence of the Facebook Forecast app is evidence against my view. I was more focused on the idea that if the recommender algorithm is based on prediction scores, that would mean that Facebook's choice of which questions to use would affect the recommendations across Facebook. 

Incentivizing forecasting via social media

I'm not an expert on social media or journalism, but just some fairly low-confidence thoughts - it seems like this is areally interesting idea, but it seems very odd to think of it as a Facebook feature (or other social media platform):

  • Facebook and social media in general don't really have an intellectual "brand". It seems likely that if you did this as a Facebook feature, it would be more likely to get dismissed as "just another silly Facebook game." Or if most of the people using it weren't putting much effort into it, the predictionslikely  wouldn't be that accurate, and that could undermine the effort to convince the public of its value.
  • The part about promoting people with high prediction scores seems awkward. Am I understanding correctly that each user is given one prediction score that applies to all their content? So that means that if someone is bad (good) at predicting COVID case counts, then if they post something else it gets down- (up-) weighted, even if the something else has nothing to do with COVID? That's likely to be perceived as very unfair. Or do you have some system to figure out which forecasting questions count toward the recommender score for which pieces of content? Even then it seems weird - if someone made bad predictions about COVID in the past, that doesn't necessarily imply that content they post now is bad.
  • Presumably the purpose of this is to teach people how to be better forecasters. If you have to hide other people's forecasts to prevent abuse, then how are you supposed to learn by watching other forecasters? Maybe the idea is that Facebook would produce content designed to teach forecasting - but that isn't the kind of content that Facebook normally produces, and I'm not sure why we would expect Facebook to be particularly good at that.
  • All the comparisons between forecasting and traditional fact-checking are weird because they seem to address different issues; forecasting doesn't seem to be a replacement or alternative to fact-checking. For instance, how would forecasting have helped to fight election misinformation? If you had a bunch of prediction questions about things like vote counts or the outcomes of court cases, by the time those questions resolved everything would be already over. (That's not a problem with forecasting, since it's not intended for those kinds of cases. But it does mean that  it would not be possible to pitch this as an alternative to traditional fact-checking.)
  • In general, this seems to require a lot of editorial judgment on the part of Facebook as to what forecasting questions to use and what resolution criteria. (Especially this would be an issue if you were to use a user's general forecasting score as part of the recommender algorithm - for instance, if Facebook included lots of forecasting questions about economic data, that would end up advantaging content posted by people who are interested in economics, while if the forecasting questions were about scientific discoveries instead, then it would instead advantage content posted by people who are interested in science.) My guess is that this sort of editorial role is not something that social media platforms would be particularly enthusiastic about - they were sort of forced into it by the misinformation problem, but in that case they mostly defer to reputable sources to adjudicate claims. While they could defer to reputable sources to resolve questions, I'm not sure who they would defer to to decide what questions to set up. (I'm assuming here that the platform is the one setting up the questions - is that the case?)
  • Another way to game the system that you didn't mention here: set up a bunch of accounts, make different predictions on each of them, and then abandon all the ones that got low scores, and start posting the stuff you want on the account that got a high score.

 

I wonder if it might make more sense to think of this as a feature on a website like FiveThirtyEight that already has an audience that's interested in probabilistic predictions and models. You could have a regular feature similar to The Riddler but for forecasting questions - each column could have several questions, you could have readers write in to make forecasts and explain their reasoning, and then publish the reasoning of the people who ended up most accurate, along with commentary.

[Linkpost] Some Thoughts on Effective Altruism

You mention that:

Neither we nor they had any way of forecasting or quantifying the possible impact of [Extinction Rebellion]

and go on to talk about this is an example of the type of intervention that EA is likely to miss due to lack of quantifiability.

One think that would help us understand your point is to answer the following question:

If it's really not possible to make any kind of forecast about the impact of grassroots activism (or whatever intervention you would prefer), then on what basis do you support your claim that supporting grassroots activism would improve its impact? And how would you have any idea which groups or which forms of activism to fund, if there's no possible way of forecasting which ones will work?

I think the inferential gap here is that (we think that) you are advocating for an alternative way of justifying [the claim that a given intervention is impactful] other than the traditional "scientific" and "objective" tools (e.g. cost-benefit analysis, RCTs) , but we're not really sure what you think that alternative justification would look like or why it would push you towards grassroots activism.

I suspect that you might be using words like "scientific", "objective", and "rational" in a narrower sense than EAs think of them. For instance, EAs don't believe that "rationality" means "don't accept any idea that is not backed by clear scientific evidence," because we're aware that often the evidence is incomplete, but we have to make a decision anyway. What a "rational" person would say in that situation is something more like "think about what we would expect to see in a world where the idea is true compared to what we would expect to see if it were false, see which is closer to what we do see, and possibly also look at how similar things have turned out in the past."

[Linkpost] Some Thoughts on Effective Altruism

A more charitable interpretation of the author's point might be something like the following:

(1) Since EAs look at quantitative factors like the expected number of lives saved by an intervention, they need to be able to quantify their uncertainty.

(2) Interventions that target large, interconnected systems are harder to quantify the results of than interventions that target individuals. For instance, consider health-improving interventions. The intervention "give medication X to people who have condition Y" is easy to test with an RCT. However, the intervention "change the culture to make outdoor exercise seem more attractive" is much harder to test: it's harder to target cultural change to a particular area (and thus it's harder to do a well-controlled study), and the causal pathways are a lot more complex (e.g. it's not just that people get more exercise, it might also encourage changes in land-use patterns, which would affect traffic and pollution, etc.) so it would be harder to identify what was due to the change.

(3) Thus, EA approaches that focus on quantifying uncertainty are likely to miss interventions targeted at systems. Since most of our biggest problems are caused by large systems, EA will miss the highest-impact interventions.

[Linkpost] Some Thoughts on Effective Altruism

As for the question of "what do the authors consider to be root causes," here's my reading of the article. Consider the case of factory farming. Probably all of us agree that the following are all necessary causes:

(1) There's lots of demand for meat.

(2) Factory farming is currently the technology that can produce meat most efficiently and cost-effectively.

(3) Producers of meat just care about production efficiency and cost-effectiveness, not animal suffering.

I suspect you and other EAs focus on item (2) when you are talking about "root causes." In this case, you are correct that creating cheap plant-based meat alternatives will solve (2). However, I suspect the authors of this article think of (3) as the root cause. They likely think that if meat producers cared more about animal suffering, then they would stop doing factory farming or invest in alternatives on their own, and philanthropists wouldn't need to support them. They write:

if all investment was directed in a responsible way towards plant-based alternatives, and towards safe AI, would we need philanthropy at all

Furthermore, they think that since the cause of (3) is a focus on cost-effectiveness (in the sense of minimizing cost per pound of meat produced), then focusing on cost-effectiveness (in the sense of minimizing cost per life saved, or whatever) in philanthropy promotes more cost-effectiveness focused thinking, which makes (3) worse. And they think lots of problems have something like (3) as a root cause. This is what they mean when they talk about "values of the old system" in this quote:

By asking these questions, EA seems to unquestioningly replicate the values of the old system: efficiency and cost-effectiveness, growth/scale, linearity, science and objectivity, individualism, and decision-making by experts/elites.

As for the other quote you pulled out:

[W]ealthy EA donors [do] not [go] through a (potentially painful) personal development process to confront and come to terms with the origins of their wealth and privilege: the racial, class, and gender biases that are at the root of a productive system that has provided them with financial wealth, and their (often inadvertent) role in maintaining such systems of exploitation and oppression.

and the following discussion:

To be more concrete, I suspect what they're talking about is something like the following. Consider a potential philanthropist like Jeff Bezos - they likely believe that Amazon has harmed the world through their business practices. Let's say Jeff Bezos wanted to spend $10 billion of his wealth on philanthropy. There might be two ways of doing that:

(1) Donate $10 billion to worthy causes.

(2) Change Amazon's business practices such that he makes $10 billion less money, but Amazon has a more positive (or less negative) impact on the world.

My reading is that the authors believe (2) would be of higher value, but Bezos (and others like him) would be biased toward (1) for self-serving reasons: Bezos would get more direct credit for doing (1) than (2), and Bezos would be biased toward underestimating how bad Amazon's business practices are for the world.

---

Overall, though I agree with you that if my interpretation accurately describes the author's viewpoint, the article does not do a good job arguing for that. But I'm not really sure about the relevance of your statement:

My impression is there's a worldview difference between people who think it's possible in principle to make decisions under uncertainty, and people who think it's not. I don't have much to say in defense of the former position except to vaguely gesture in the direction of Phil Tetlock and the proven track record of some people's ability to forecast uncertain outcomes.

Do you think that the article reflects a viewpoint that it's not possible to make decisions under uncertainty? I didn't get that from the article; one of their main points is that it's important to try things even if success is uncertain.