Enjoyed the episode? Want to listen later? Subscribe by searching 80,000 Hours wherever you get your podcasts, or click one of the buttons below:

I think it’s certainly fair to say that there is more good journalism than people realize. And the reason for that is that a lot of the stuff that gets people very angry is not the best stuff out there.

Kelsey Piper

“Politics. Business. Opinion. Science. Sports. Animal welfare. Existential risks.” Is this a plausible future lineup for major news outlets?

Funded by the Rockefeller Foundation and given very little editorial direction, Vox’s Future Perfect aspires to be more or less that.

Competition in the news business creates pressure to write quick pieces on topical political issues that can drive lots of clicks with just a few hours’ work.

But according to Kelsey Piper, staff writer for this new section on Vox’s website focused on effective altruist themes, Future Perfect’s goal is to run in the opposite direction and make room for more substantive coverage that’s not tied to the news cycle.

They hope that in the long-term, talented writers from other outlets across the political spectrum, can also be attracted to tackle these topics.

Some skeptics of the project have questioned whether this general coverage of global catastrophic risks actually helps reduce them.

Kelsey responds: if you decide to dedicate your life to AI safety research, what’s the likely reaction from your family and friends? Do they think of you as someone about to join “that weird Silicon Valley apocalypse thing”? Or do they, having read about the issues widely, simply think “Oh, yeah. That seems important. I’m glad you’re working on it.”

Kelsey believes that really matters, and is determined by broader coverage of these kinds of topics.

If that’s right, is journalism a plausible pathway for doing the most good with your career, or did Kelsey just get particularly lucky? After all, journalism is a shrinking industry without an obvious revenue model to fund many writers looking into the world’s most pressing problems.

Kelsey points out that one needn’t take the risk of committing to journalism at an early age. Instead listeners can specialise in an important topic, while leaving open the option of switching into specialist journalism later on, should a great opportunity happen to present itself.

In today’s episode we discuss that path, as well as:

  • What’s the day to day life of a Vox journalist like?
  • How can good journalism get funded?
  • Are there meaningful tradeoffs between doing what’s in the interest of Vox, and doing what’s good?
  • How concerned should we be about the risk of effective altruism being perceived as partisan?
  • How well can short articles effectively communicate complicated ideas?
  • Are there alternative business models that could fund high quality journalism on a larger scale?
  • How do you approach the case for taking AI seriously to a broader audience?
  • How valuable might it be for media outlets to do Tetlock-style forecasting?
  • Is it really a good idea to heavily tax billionaires?
  • How do you avoid the pressure to get clicks?
  • How possible is it to predict which articles are going to be popular?
  • How did Kelsey build the skills necessary to work at Vox?
  • General lessons for people dealing with very difficult life circumstances

Rob is then joined by two of his colleagues – Keiran Harris and Michelle Hutchinson – to quickly discuss:

  • The risk political polarisation poses to long-termist causes
  • How should specialists keep journalism available as a career option?
  • Should we create a news aggregator that aims to make someone as well informed as possible in big-picture terms?

Get this episode by subscribing to our podcast on the world’s most pressing problems and how to solve them: type 80,000 Hours into your podcasting app. Or read the transcript below.

The 80,000 Hours Podcast is produced by Keiran Harris.


There’s very few complicated ideas where reading one 2,000 word article about it is going to stick with you as a significant change in your understanding of the world. I think what it can do is maybe get you interested enough to read more, and over time there can be lots of articles that maybe reading all of them can be a little bit more compelling, but yeah, it seems absolutely true to me that you can’t expect your case for impact to be we wrote this article, tons of people read it, they changed their minds and got a more productive understanding of things.

I think you basically can’t pitch effective altruism to everybody, and I think most people have a comparative advantage at pitching effective altruism to people who are going to find it compelling for reasons sort of like theirs. That doesn’t mean you should just write articles that would’ve convinced a past you, but it does mean that if you’re religious, you’re going to do better at explaining religion to effective altruists than if somebody who’s not religious tries to do that. If you’re coming at it as an environmentalist who really wants to work on climate change, but has maybe decided that’s not the thing to do, then lean into that and talk with other people who care about climate change about effective altruism and about climate change and stuff like that. If you’re here from the animal community, then think about what effective altruism can bring the animal community, and try and bring that as many places as you can.

If you’re a smart, young person and studying or a smart, established person in your career and you see the argument for AI risk and you think, “That’s what I should be doing, I want to do that.”

Are the people around you going, “Oh, yeah. I heard about that. That’s important. I’m glad you’re working on it,” the way I think they would react if you said, “I’m going to work on making organ donations safer”, or “I’m going to work on reducing crime or like improving outcomes in policy.”

Or are people going to be like, “That weird Silicon Valley, apocalypse cult thing.” That matters. I think to a lot of people, that matters. I think all of this is very hard to quantify, but I think if the general Vox reading, smart, wanting the world better, maybe not super informed about AI in-particular, but broadly sympathetic to efforts to handle technology safely. If they’ve heard the case for AI risk and they have this general sense, “Yes, that’s something some people should be working on,” then I think that’s good.

Forecasting is really hard, and practice seems to make a big difference in improving at it. I don’t necessarily expect these predictions to make us come out looking particularly good, but I think it’s still really important to do this if you want to set that standard for journalists in general. If you want people to take your predictions about the future seriously, you need to start saying, “Here’s how seriously you should take me. Here’s how good I am at this.”


Robert Wiblin: Hi listeners, this is the 80,000 Hours Podcast, where each week we have an unusually in-depth conversation about one of the world’s most pressing problems and how you can use your career to solve it. I’m Rob Wiblin, Director of Research at 80,000 Hours.

This week we’re going to find out what it’s like to become a journalist who strives to cover the issues that really matter.

A few quick notices.

Firstly, if you’d like to come to the Effective Altruism Global conference in San Francisco from the 21 to 23 June, applications are now open at eaglobal.org. If you like this show you’re highly likely to enjoy the people you’ll meet and the conversations you’ll have there.

Secondly, most people find out about this show from personal recommendations. So if you’re enjoying the show, or know someone who would benefit from listening to it, please send them a message so they find out that it exists.

Alternatively you can leave us a review on iTunes which also helps people find out about the show.

At the end of this episode my colleagues Michelle and Keiran come on the show again for a 20 minute chat about whether journalism is a good career and the risks of political polarisation.

Finally I’ll also link to some of my favourite articles by the journalist in this episode for you to check out, including topics like what works to stop factory farming, how to improve science grant funding, and why impact investing is probably overrated.

Alright, here’s Kelsey.

Robert Wiblin: Today I’m speaking with Kelsey Piper. Kelsey is a staff writer for Vox’s new vertical focused on effective altruist themes, including threats to humanity as a whole. She previously worked as the head of the writing team at Triplebyte and ran Stanford effective altruism during college. She’s also blogged at the Unit of Caring, her own blog, for many years.

Robert Wiblin: So, thanks for coming on the podcast Kelsey.

Kelsey Piper: Thanks so much. I actually love this podcast so it’s a treat to get to be here.

Robert Wiblin: Yeah. So I’ve been really enjoying your writing since you joined Vox in last October right?

Kelsey Piper: Yeah.

Robert Wiblin: Yeah, so I hope to get to talk about careers in journalism in general. Like how we can improve the information environment out there on issues that we care about. And I suppose also some specific articles that you’ve written. But first, tell us, what are you working on at the moment? And why do you think it’s really important?

Kelsey Piper: Yes. So I report for Future Perfect. I write mostly about factory farming and animal issues, about existential risks, and about some intriguing ways to improve the world. For example, improve scientific research or improve how we do voting, some of the stuff that you guys have covered on 80,000 hours actually. These big ideas that I think aren’t necessarily getting as much air as they possibly should.

Intro to Future Perfect

Robert Wiblin: Yeah, guess it’s pretty similar to what I do in a way. Yeah, so this new section of Vox’s website Future Perfect. Tell us a little bit about the history there and how people have been reacting to it so far.

Kelsey Piper: Yes. So a Future Perfect is funded by the Rockefeller Foundation. They did not give us very much editorial direction, which I think is really good. You want journalism to not be following a set of instructions about what cover. That can sort of constrain the ability to tell the most important stories. So they were very flexible. They just wanted to create room for more coverage that’s not tied to the news cycle. That’s focused on important issues.

Kelsey Piper: A lot of the people at Vox, in particular, Dylan Matthews and Ezra Klein, were both very sympathetic to effective altruism. Both had a lot of overlap in their interest with effective altruism and had written before about those topics. They wanted Future Perfect to be heavily inspired by effective altruism and to draw on a lot of the work the effective altruism community is done on those important questions.

Robert Wiblin: What kinds of articles have you been writing about or what kind of topics have you been writing about so far and how have readers responded?

Kelsey Piper: Yes. So I cover factory farming. I cover AI and the big breakthroughs on the AI capabilities front and the case for AI safety and the way researchers are thinking about safety today. I’m writing a piece right now about bio risk and gain a function research and some of the ways that the research we’re doing right now and bio could be dangerous if we don’t handle it appropriately. Then I’ve written about improving the grant process, improving the voting process, stuff like that.

Robert Wiblin: And what’s the reception been to the to the articles? Are getting like many people reading them and do people send you emails, either like love mail or hate mail?

Kelsey Piper: Yeah. So Future Perfect has gotten an encouraging response. We’ve had a lot of people reading and engaging with the articles. We’ve had a lot of people subscribe to our newsletter which is something we look for as a sign that they’re valuing Future Perfect content, in particular, on seeking out our perspective. I don’t get very much feedback in emails. This is actually something I think people should know they could do more. Journalists tend to really value feedback and responses, especially feedback and responses that make it clear you read the whole article and thought about its content. You do get a lot of reactions that are maybe a little bit shallower responses to the headline, that’s a little bit less valuable. But I think if you see high-quality reporting and you want to let a reporter know that you valued it, that you shared it with your friends, that it taught you something? I think that we’re not nearly as drowned in that kind of feedback is I think people might anticipate. It tends to encourage more coverage with whatever it is you’re looking for.

Robert Wiblin: Yeah. Interesting. I follow Matt Yglesias, one of the Vox’s journalists, on Twitter semi-regularly. I guess, he’s like quite a public figure.

Kelsey Piper: Yes, I think he gets some more hate mail and probably more fan mail.

Robert Wiblin: Yeah. He semi-regularly posts the most horrific emails that he receives but I guess that’s pretty not typical. I guess he’s been at this for a lot of years, so he’s able to take it with a great sense of humor at this point.

Kelsey Piper: Yeah, I think I have not had that experience and I’m somewhat glad. Maybe, maybe I’ll develop a thick skin for it over time.

Robert Wiblin: Yes, so how does Future Perfect view its relationship with effective altruism as a kind of intellectual community?

Kelsey Piper: Yes. So, I think Future Perfect definitely hopes that we’re creating content that effective altruists will find it worth their time to read. I think we’re definitely hoping that a lot of our content will advance conversations in effective altruism or bring them to a wider audience. We’re not an effective altruist outlet. Future Perfect wants to also cover other things that are part of our mission but not necessarily good priorities for EA. But certainly, I think we want the EA community to be getting a lot of value from Future Perfect and I joined Future Perfect for EA reasons. I thought it was one of the best ways I couldn’t improve the world.

Robert Wiblin: Yeah. How many of you are there? It sounds like there might only be two or three.

Kelsey Piper: Right now, me and Dylan are writing for Future Perfect and we’re also in the process of hiring a community manager. On Monday, our new writer Seagal Samuels, who just left the Atlantic, is starting. So, then we’ll have another team member.

Robert Wiblin: Yeah, because Vox has a whole has 50 staff, 100 staff? It’s recently big now, right?

Kelsey Piper: It is pretty big. That’s a good question because there’s also all the sister sites. I don’t have a number for you.

Robert Wiblin: I guess I’m curious to know what does the rest of the organization think of Future Perfect and your work. They’re kind of a little bit more focusing on the issues of the day and politics and so on. I wonder, like whether they think this is the wonky, like academic side of Vox.

Kelsey Piper: So far the reaction has been really positive and really encouraging. We’ve had a lot of writers who are interested in when one of their stories might fit within Future Perfect, they’re excited to have it published there. They like that we’re doing that and are excited to contribute that. We’ve gotten a lot of advice.

Kelsey Piper: The science and health team obviously contribute a lot of articles to Future Perfect when something in their purview comes up, the team that works on global warming. So I think we really couldn’t do something like Future Perfect without the rest of Vox just because you need a lot of science expertise to do good science reporting. You need a lot of climate change expertise to do good climate change reporting. And Dylan and I don’t have that, but we have this great team we can draw on.

Which FP articles are you most proud of?

Robert Wiblin: Yes, which articles that Future Perfect has put out so far are you kind of most proud of? Are there anywhere you feel it’s actually moved the needle and make the world a better place? It’s early days, it’s only been around for five months or, maybe less than that, three months, four months?

Kelsey Piper: Yeah, we launched on October 15. So, I think it’s a little early to start seeing if we’re influencing the world, but I’m really proud to see factory farming coverage be a little bit more mainstream.

Kelsey Piper: It’s just something that seems like it should be covered along with all of the other issues of the day and excited to see it fit into that role. I was really happy when I published a piece about AI safety a while back. I heard from some people that that gave them a clear explanation of what was going on with AI to point to on. That’s something that seems potentially pretty valuable, just to make sure people are on the same page about that.

Robert Wiblin: Yeah, I think you wrote an article about the state of evidence and animal advocacy. So yeah, do we have evidence for like leafleting in favor of vegetarianism or veganism working and online ads and things like that. Do you wanna describe what you found there?

Kelsey Piper: Yes, so I really just had conversations with a bunch of people who are doing this research in the animal field. That’s one of the best things about being a journalist, is that I can call up experts in the field and say, just tell me what’s going on and have great conversations. That’s a perk I didn’t really anticipate that people be so willing to talk with me. But what I heard from several of them.

Kelsey Piper: I heard from the team at Faunalytics, I heard from the team at the ACE or Animal Charity Evaluators, was we’re seeing these corporate campaigns be hugely successful. We’re not seeing very much of an evidence base for a lot of things that have traditionally been a focus of these animal groups. In particular, leafleting but also more generally, any efforts to kind of convince the public to become vegetarian or vegan. So it was like, I feel like there’s lots of people who care about animals and care about factory farming who aren’t aware of where the evidence is at. And haven’t seen the case that corporate campaigns are where to focus and so got to write about that.

Robert Wiblin: Yeah, I just brought up that article because I just thought it was the best thing that I’d seen written on that topic. Like evaluating what things do we have evidence is working and what things don’t we and summarizing it all like very quickly. I thought like I was really learning a lot and I kind of slightly work in this area. So it’s great that they can be an article in the mainstream media where I feel like I’m getting seriously informed.

Kelsey Piper: Wow. Thank you.

Day-to-day life at Vox

Robert Wiblin: Yeah. So, what is kind of the day to day life of a journalist like? I imagine it’s kind of hectic. The demands to put out content a pretty serious, I’ve heard.

Kelsey Piper: Yeah, Vox has a very fast pace which was definitely something I was a little apprehensive about going in, like can I write that much? But it’s been very good for me because I think the push to think about something you want to tell people every day just keeps you moving. On most days I will try and send my editor about three story ideas. Things that I’ve thought of that I want to write about, things that I have a lead on, things that I saw in the news that I felt like we needed a Future Perfect take on. My editor will get back to me with the one or two that he’s most excited about and say, Yeah, go ahead and write this story.

Kelsey Piper: So, then I’ll email people who I want to talk to. I’ll try and get introductions. I’ll research for the piece. I’ll have those conversations and phone calls. I’ll try and write the piece. I’ll try and file it before I go home. Then often, at the same, time my editor and I will be going back and forth with edits on yesterday’s story to get it to a state where we’re both proud of it and confident of it and ready to put it on the site.

Robert Wiblin: OK, so now on a typical day you have like two things on the boil. One that you are starting today and one that are finishing from yesterday. And the goal is to hopefully publish something basically every workday?

Kelsey Piper: Yes. Now, in practice, some pieces take longer to come together. Or they come partway together and then we realize there’s not a good story here. Or the situation is confusing enough that our initial take on it didn’t work. A fair number of stories get scrapped. In practice, I think I end up publishing four things a week. But yeah, the goal is certainly to have a week where every day we put out a new story.

Robert Wiblin: Okay, so you arrive in the office on Monday morning. The first thing you have to do is figure out what you’re going to write about that day? How do you go about that?

Kelsey Piper: So, I do a couple different things. One is I keep an eye out on the EA forums, on the EA discord, in the Facebook groups, just for things people seem to be confused about. Those often make for great stories. Or things that I’m a little confused about and would love to just spend four hours digging into until I have a clearer picture of what’s going on.

Kelsey Piper: And then I also look at the news. We are covering a fair bit of philanthropy, is another thing that EA is not necessarily focused on, that Future Perfect is pretty interested in. It’s like coverage of the big philanthropists. What is Bill Gates doing? What is Jeff Bezos doing? So I’ll check the news and look for stories that seem like there’s a lot of takes out there but it would be valuable to have a sort of Future Perfect take out there. Then I’ll look at research that just came out especially research in development economics or in health interventions that give well supports or on other topics that are of interest.

Kelsey Piper: A write up a study is always a great piece because it’s pretty straightforward. The authors are usually happy to talk with you and make sure you understand their research. Then you just explain the study, explain how confident we should be in it. Stuff like that.

Robert Wiblin: Okay, so, at the start of the day you come up with, was it five or ten ideas for articles to write?

Kelsey Piper: I aim for three.

Robert Wiblin: Three. Okay. And then, how do you choose among those?

Kelsey Piper: My editor will usually take a look. M editor has better instincts than me for what will our Vox audience like and which of these are going to turn into a solid story.

Robert Wiblin: And then how do you go about writing it? I mean most people I think would find it quite hard to write an article within a day or two. Most people find it very hard to put pen to page to begin with.

Kelsey Piper: Yeah, I think Vox has a big focus on explaining things, on answering questions. So you want to put yourself in the mind of an audience and ask, “What questions are they going to have here?” And often, you have to start by explaining to them, why do they care about this. Why are you telling them about this? Then they’re going to wonder, why aren’t people doing this obvious solution I thought. I heard something about this, why doesn’t that work?

Kelsey Piper: So I think often articles are sort of shaped around figuring out where the readers at and then telling them, this is interesting. Here’s some things you’re wondering about it. Here’s the answer. Then here’s a takeaway, our understanding of what’s going on. I like that style of writing because it’s very audience focused. It seems very suited to a lot of the topics I’m interested in. Where you both want to engage people and explain why they care. And give them a somewhat complicated picture of what’s going on with some takeaways that hopefully they can use to make better decisions and focus on the stuff that’s important.

Robert Wiblin: So I mentioned, if I had to write something every day that I would often get to the mid-afternoon be like, I don’t know what I think about this issue. I haven’t really figured out what the answer is yet. Or am I get there and be like, Oh, wow, my whole like take on this was just…I misunderstood the whole issue. And now it’s like 3pm and I’ve got to file something I guess, within the next few hours. But I don’t know what to say at this point.

Kelsey Piper: I get a pretty good reaction when I say, sorry, the story is going to take another day because it turned it into a different story. Sometimes that means we scrap the story. Sometimes, it means I write one that has a different, more complicated take. I think there’s not much pressure. Aside from, your desire to file the story and get to stop to write something that you’re starting to feel like is more complicated than that. Because complicated is okay and the journey you had in the process of figuring it out is probably a journey you want to take the readers through. It’s in some ways more interesting than whatever your original take was.

Robert Wiblin: Yeah, okay. So if you decide that actually, things are quite different than what you originally thought then you can just explain the process by which he got there. And that’s like an interesting story in its own right. You didn’t have to stick with the original vision, just to like get something out there.

Kelsey Piper: Yeah, sometimes that’s a compelling story in its own right. And sometimes it’s not, and the story ends up getting scrapped. But both of those are fine outcomes. You don’t want to put something out there that you’re feeling, even as you post it like I don’t know how much I stand behind that. That’s just not gonna to be reporting you can really stand by in the long run.

Tradeoffs between the interests of Vox and doing the most good?

Robert Wiblin: Do you find that there are meaningful tradeoffs between just doing what’s in the interest of Vox as a business or what is it good? in terms of building your brand as a journalist? What seems to be like most effective from an EA point of view?

Kelsey Piper: Yes, so I think a lot of journalists run into the same conundrum here. Which is that takes on the news of the day, political stuff, polarizing sort of takes, tend to do very well. If you’re looking at page views, those will get overwhelmingly good page views. I think the article I wrote that did the best was the one that was titled “Billionaires Don’t Run for President”. I stand by the article. I feel like it’s a pretty good explanation of why, if you’re a billionaire and you have some goals, you have better avenues to pursue those goals. The evidence that self-funded candidates do well and the road for office is pretty mixed. So I feel like it’s a good article. But certainly that headline “Billionaires Don’t Run for Office” the timing, the day of Howard Schultz announcement, that made that like a really exciting story that everybody wanted to share. Compared to something that’s just an explanation of the literature on skills development programs in the developing world, it isn’t going to get that kind of response.

Kelsey Piper: So, I think in Future Perfect but probably less in Future Perfect than in journalism in general, there’s certainly a feeling of like what I want people to read the things I write. And if I write about politics, and if I have these provocative headlines and if I go for the sort of more interesting takes, then everybody reads what I write. And if I try and step back the rhetoric, and if I try and add nuance, and if I try and you know tell a more complicated story that no one really wants to hear, then no one will hear it.

Kelsey Piper: I think that’s I see a lot of takes on why journalism has gotten or seems to many people to have gotten more polarizing and more driven by anti-Trump, pro-Trump, back and forth. I actually don’t think it’s journalists really wanting to convince the public to be pro-Trump or anti-Trump. I think it’s that the articles that aren’t about that don’t get shared and don’t get liked. That both makes it hard as a journalist to tell the stories that you think it’s most important to tell. And it means that as an audience member the stories you see your friends sharing will all be selected from that subset, that’s sort of more provocative. Future Perfect is somewhat insulated from this because we don’t have page view goals. Because while politics is sometimes within our purview, certainly what did Trump say today is not going to be within our purview.

Kelsey Piper: And because we have this mission we can hide behind. It’s like we’re trying to tell the big picture. But I think lots of journalists who would love to be doing substantive reporting are in hard place if they also want people to read what they’re writing.

Robert Wiblin: Yeah. So how do you reconcile those goals? I don’t suppose, you’re someone insulated from this. You have the good fortune as I do, doing the show, to just go into great detail about things that we actually think are important. But what should someone who’s in a less fortunate position do? Is there any way of squaring the circle?

Kelsey Piper: I think in the long run there’s certainly still a market for nuance and complexity and good reporting. I know a lot of people who do very good stuff in that vein. I think there are also a lot of people who’ve chosen the sort of tradeoff of writing about the latest Trump tweet but trying to do that in a way that paints a nuanced picture of a complicated subject. Using the stories of the day as the hook into that. That seems to me like the valid endeavor. Tell people the stories that they’re interested in reading about and then try and tell those stories with depth and nuance and complexity and just getting them right.

Robert Wiblin: So kind of seems on this story that, in terms of who’s to blame for a lot of journalism being quite bad, it’s the audience. Because they’re they’re choosing to read and share articles that are not of great lasting value. And so, to some extent, we’ve seen the enemy and its ourself. Is that fair to say?

Kelsey Piper: I think it’s certainly fair to say that there is more good journalism out there than people realize. And the reason for that is that a lot of the stuff that gets people very angry is not the best stuff out there. And not even the best stuff that those writers are producing and writing. I think what’s going on with journalism as an industry is very complicated. I’ve not been involved with it for long enough to be very confident in it. I think local news is dying in a way that’s very bad for communities. Because it’s often the main check on corruption and bad local politics and abuses of power on the local level and stuff like that. It’s dying in part because people actually prefer to read national news. So that one you can say is it’s following the consumers and the consumers have a bit of a collective action problem in the sense that they might care about like corruption being thwarted but they don’t care to read the local paper.

Kelsey Piper: I think digital media is been in the headlines in the last couple of weeks with a lot of major layoffs and job cuts that has a bunch of people saying, “Is digital media a good idea?” I think that’s just that digital media was pitched as a startup, like tech. Like you become big, you become a monopoly, you have these huge margins. People eventually realized it’s not going to look like that. It’s not going to be a huge margin monopoly business. It’s going to be a low barrier to entry competitive business. That meant investor expectations and funding levels were sort of out of line with the business models that made any sense. I don’t think it means digital media is over. I think it just means it’s going to be a low margins industry that’s going to stick around like that.

Kelsey Piper: And then there’s the polarisation scheme. I think there’s just a lot of things going on there. It would be very complicated to change any of them. Because a lot of them are just market forces acting quite strongly.

Robert Wiblin: Yeah, I guess we’ll return to journalism as an industry later on. So, when I visit one of your articles on Vox, the amount of revenue that Vox gets is probably like a fact tiny fraction of a cent. It actually maybe zero. Maybe it’s negative because I have an ad blocker on my browser. So I suppose there’s no point making any content for me. People should just avoid writing articles to me and people like me, who tend to have ad blockers on because it’s just there’s nothing to be gained. Is that right?

Kelsey Piper: I don’t know very much about Vox’s revenue model but that doesn’t seem far off to me. I do think that adds driven companies like Vox are in a different situation than subscription driven ones like the New York Times and The Washington Post moved to. You need lots of viewers, you need lots of viewers who are watching and hopefully clicking on ads. A lot of people just aren’t.

Robert Wiblin: Yeah, I have been wondering, so I think at the moment I don’t subscribe to any newspapers. In the past, I’ve had subscriptions to places that produce content. But lately, I haven’t. I’ve been feeling bad about that. Like maybe I should just always have a subscription to whatever place I think is doing the best work. So I can kind of do my fair share to contribute to actually creating content that keeps politics sane or keeps where we’re able to learn really useful things, is kind of contributing to a public good.

Kelsey Piper: I do think that paying for content you value is, obviously not on the individual scale but on the larger scale, the only way to expect the world in five years to have content you value happening. Which is tough because a lot of EA’s, where their money goes is already a really important question. I don’t necessarily think that they should be spending it. But I do think that as a society if we don’t pay for good reporting, then there’s going to be-

Robert Wiblin: It will not be there.

Kelsey Piper: Yeah.

Risk of political polarisation

Robert Wiblin: Yeah. So some people have been concerned that Future Perfect, though it’s not kind of a specifically an effective altruism thing, it’s kind of effective altruism inspired, and effective altruism associated, and it’s probably the only EA inspired live section on a newspaper or a magazine, and I suppose that this could lead to effective altruism being associated with liberalism, because it’s in Vox, and Vox is well-known as having a general that kind of center left, liberal attitude. So do you think it would be good if more different places, particularly across the political spectrum, had sections that had a kind of effective altruism flavor, so there wasn’t this risk of EA being seen as partisan?

Kelsey Piper: I think I have complicated feelings there. So, one thing is that I think effective altruism is, it’s a couple different things. It’s a movement, it’s like a philosophy and an approach to answering questions, it’s a bunch of specific resources for people who are interested in working on specific problems. Some of those problems are likely to end up being part of political coalitions of one sort or another, and some of them probably are much better off if they aren’t sort of framed in those terms at all. I think it’s really unclear to me whether effective altruism being associated with center left politics is a problem to mitigate, an inevitable consequence of the sort of issues that effective altruism is about. Like, I don’t know if you could get The Wall Street Journal to write about factory farming, even if you offered them a grant to do that. I’m not sure whether there are people who want to work at The Wall Street Journal and want to write about factory farming.

Kelsey Piper: I worry a little bit about targeting perceptions of where EA is on the political spectrum instead of just doing what I think Future Perfect is trying to do, which is more just borrow some of the ideas of EA and write content that is valuable to EAs, but avoid that branding and therefore keep the EA movement hopefully can be branded however it wants. Future Perfect is Vox’s thing.

Kelsey Piper: That said, there are a bunch of specific issues that Future Perfect covers that I do think should just be covered everywhere. I think policy with respect to existential risk should just be an area that newspapers cover because it’s important. I think international development and health is, to some extent already, but should be more so just an issue that all newspapers should cover. I think it would certainly be a really good sign if newspapers across the political spectrum were covering factory farming as a problem. Yeah, there are a lot of issues where I would be very excited to see other people picking up Future Perfect’s manner.

Kelsey Piper: Then there are lot of issues where I think it would really matter who they hired. I feel like Future Perfect is something I’m so excited about in large part because Dylan is the person who sort of drove it, and Dylan is somebody who cares a lot about effective altruism, and cares a lot about the causes and priorities I care about, and because Future Perfect’s team was so open to hiring people with a background in the effective altruism community to sort of lend that perspective. I think if you tried to have it, for example, an AI department, and you didn’t specifically hire for background in the effective altruism community, then I’m not sure I expect that AI department to do anything good, so I would have some reservations about trying to convince a Future Perfect without a clear picture of what are the internal institutional incentives there, who would be writing the stories, and what would our goal be, even though I think many of these stories should be covered by everybody because they’re important.

Robert Wiblin: Yeah, it is interesting. You think, I guess, like the three cause areas most associated with effective altruism, you got kind of global health, animal welfare, and global catastrophic risks. It feels like global catastrophic risks just clearly shouldn’t … really isn’t that partisan at the moment, or at least in principle I don’t think there’s Republicans who are in favor of nuclear war.

Kelsey Piper: Yeah, no, I think almost everybody expresses concerns about nuclear war, and in fact I’ve talked to a fair number of conservatives who specifically supported Trump because they thought he was less likely to have a dangerously adversarial relationship with Clinton, and they thought that Clinton’s approach was more likely to lead to nuclear war, and certainly it seems good to me for there to be more national conversation about which of these politicians’ approach to geopolitics endangers the world more.

Kelsey Piper: Factory farming seems likely to continue to be somewhat associated with liberals, but I’m not sure that’s necessary. I’ve seen a lot of comparisons to gay rights, which were fringe, and then they were mainstream among Democrats, and then that was enough to sort of get wide spread successes without them ever being mainstream among Republicans in the US. But, that has a bunch of religious implications. Republicans in the US are religious. Factory farming doesn’t seem intractable among Republicans in the same way to me. It seems like it might actually be on the left side of all of this just because it is associated with that and those are the people pushing for it, so I would certainly be excited about, like I said, I’d be thrilled if The Wall Street Journal had a factory farming beat. I would be thrilled if some of the more conservative outlets that are like the new media outlets of whatever had a factory farming beat. That would just straightforwardly be great news.

Robert Wiblin: Yeah, it’s interesting. It feels like the animal welfare thing is perhaps rather than like a right, left thing, it’s more of an urban, rural, like the urban rural divide. It probably looms larger, though I think even there if you ask people in principle, are they against causing suffering to animals on farms, just the vast majority of people in every group are against this. I’m not quite sure why it is the case that it’s like that these issues are just so much more prominent among liberal left groups. I guess it’s possible because they’re much more urban in general, so it conflicts less with people’s livelihoods.

Kelsey Piper: And younger people seem more concerned about animal issues.

Robert Wiblin: I will say overseas that’s much less the case. In the UK, there isn’t really a partisan, or there isn’t really a left, right split going on in animal welfare. In fact, the Conservative party has pushed forward on animal welfare reforms, I think, just as much as parties on the left have. And in Australia, I think it is a bit more like it is here, but nonetheless, I think there’s not particular reason why it has to be partisan, and I think that it’s totally foreseeable that it might be much less so in the future.

Kelsey Piper: Huh. With that context, yeah, that does make me think, okay, trying to mitigate the extent to which animal welfare is seen as a partisan issue in the US is pretty important for the animal welfare movement, and if Vox contributed to making it seem like a liberal issue, then that would be unfortunate, because I think it’s just a universal issue, and I, yeah, want coverage to reflect that.

Robert Wiblin: Yeah, I wonder if it’s possible to write articles that kind of have a spin of yeah, people on … or treating people who are conservative who might otherwise not be treated in such a flattering light in Vox, like finding conservatives who are very worried about animal welfare and presenting their take on things.

Kelsey Piper: Yeah, that’s interesting. I can definitely take a look at what’s out there.

Robert Wiblin: Yeah. In terms, I guess, on global health and development, it’s interesting. That’s another case where I think the UK, there really isn’t such a big divide between the left and right, I think, on questions of international development and poverty alleviation, but again, in Australia and the US, it does seem to be more of an issue that is brought up far more on the left, and there’s some greater hostility among more conservative parties.

Kelsey Piper: So, keep in mind my co-worker, Dylan Matthews, wrote this great article arguing that George Bush’s work on AIDS in Africa was not just the biggest deal of his presidency, but was actually … he saved more lives there than he destroyed with the pointless wars in the Middle East. Which, I don’t think we need to add those up and say okay, he was good overall, but it’s still an important perspective to have, that this huge deal, the most important thing about his presidency wasn’t really getting a lot of coverage or a lot of interest, and I deeply respect what he did there.

Kelsey Piper: The current Republican party seems a bit more nationalist, a bit more hostile to the idea of significant investment in people overseas to some extent. I know Trump has a lot of rhetoric about how we’re being cheated, whenever we’re giving things to other countries and getting nothing in return, so under this administration it seems a little trickier to hit the bipartisan balance, but again, yeah, it doesn’t feel inevitable to me. It seems like an issue that certainly everybody cares about and that is possible to have investment in from all sides.

Robert Wiblin: Yeah, so let’s back up a little bit. It’s a very tricky question, but how much do you think effective altruism as a community should strive to not be seen as partisan? I guess you were saying you’re not sure that people should make active, specific efforts to avoid being seen as more liberal. Maybe you just think we should make the arguments as they are, and then the cards fall as they may in terms of what the partisan lean is.

Kelsey Piper: Yeah. I think a lot of the discussion I see effective altruists engage in around whether they should be partisan seems focused on things that would be pretty costly to change, and I would be excited to instead see people, for example, make the pitch for a cause they care about in terms they expect to be compelling to audiences they think we’re under serving, and if that results in lots more conservatives hearing about it, then that’s great. I would be intrigued by efforts to sort of make effective altruism more compelling from other perspectives. I’ve written a little bit on my Tumblr, for example, about what the best approach to reducing deaths of fetuses might be from the perspective of pro-life people, and I would be pretty excited about there being effective altruists who are interested in that question the same way there are effective altruists interested in lots of questions that might assume some values I don’t share. That seems like all good stuff that I would be excited about.

Kelsey Piper: I think a lot of the things I see people suggesting to get EA less political seem more like don’t engage in politics or don’t have discussion of which candidates are going to have the most beneficial impact on the world, and I’m less optimistic about that. That seems like it compromises our ability to do stuff by quite a lot.

Robert Wiblin: Yeah, it does seem it’s hard sell to gag people from saying things that they think are true and important. It seems like it’s easier just to add additional voices, say I’m going to write something about animal welfare in The Wall Street Journal or the National Review rather than try to get other people to shut up.

Kelsey Piper: Yeah, I think that’s a great summary. I would love it if effective altruism included more conversations and for other audiences. I get nervous when it seems like it’s should we restrict conversations for these audiences because, yeah.

Robert Wiblin: I guess I would think it was quite foolish if someone was trying to portray a global catastrophic risk as a left or liberal issue, because it’s like wrong, but I guess I haven’t seen that, and I’m not sure what the angle there is. I suppose maybe you’re criticized for right wing foreign policy or something, say oh evident these people don’t care about civilization or stability, but I think that would be a misunderstanding.

Kelsey Piper: Yeah, I think certainly for any issue, alienating lots of the country by asserting they don’t care about it looks like a pretty bad move, especially something like the destruction of civilization, which I think it appears to be the case that almost everybody wants to prevent that.

Robert Wiblin: Yeah. We could be at least that charitable. This might be a hard question, but are there any other media outlets, or what would be another media outlet that you’d be most excited about starting another kind of effective altruism inspired vertical, or section I suppose, as normal people call it?

Kelsey Piper: Again, I think that depends so much on the internal support for the division and who gets to work there. I think, obviously, the outlet and the platform matters, but almost more important to that I think is what sort of direction they’re getting, and who they are, and how much background they have in covering these issues, who they know to talk to. I think even the best reporter in the world is going to have a hard time with accurate coverage of a beat they just don’t have any exposure to in the past, and I think even the best reporter in the world is going to end up sort of sidelined into the things that their editors understand, if their editors aren’t coming at it from a perspective of really being willing to step outside their comfort zone and do something new, so I feel like lucky that Vox had editors Ezra and Dylan. I feel like that’s pretty key to Future Perfect’s being able to do what it’s doing, and I’m sure there are more Ezras and Dylans out there, but I don’t know where.

The fidelity model of spreading ideas

Robert Wiblin: So, an idea that’s become popular in the effective altruism community over the years is that in order to impact, we really want to present our views in a very sophisticated version, and to get people to really understand them on a deep level, and that there’s been experience that when they get simplified in order to get promoted, very often so much of the subtly is lost, that people can’t meaningfully act on it, that they just get kind of this garbled version of it. I think there’s a [inaudible 01:17:09] say that the Center for Effective Altruism put out the fidelity model of spreading ideas, which kind of makes this case that we really want to find mediums like long form podcasting where people can really actually grapple with the idea. I guess, yeah, do you worry it’s possible that the articles of the length that Future Perfect puts out just might not be quite in depth enough for people to fully kind of grock effective altruism?

Kelsey Piper: I think yeah. There’s very few complicated ideas where reading one 2,000 word article about it is going to stick with you as a significant change in your understanding of the world. I think what it can do is maybe get you interested enough to read more, and over time there can be lots of articles that maybe reading all of them can be a little bit more compelling, but yeah, it seems absolutely true to me that you can’t expect your case for impact to be we wrote this article, tons of people read it, they changed their minds and got a more productive understanding of things.

Robert Wiblin: Yeah. I guess you think as a package, if they read many of these articles, then maybe they’ll become more informed and more able to achieve good in their life.

Kelsey Piper: I think that people can click through and interact with other content and find the in depth explanations and find the experts, and hopefully the people who are gripped by the initial idea are willing to sort of take those follow up steps.

Robert Wiblin: Yeah, what would happen if you are someone else went into Future Perfect and said, “I think it would be good if we wrote half as many articles, but that they were twice as long.” So they went into an unusual depth on the topics. Would that be something that Vox would be excited about?

Kelsey Piper: I think Vox believes pretty strongly that having content go up frequently is pretty essential to engagement, and that telling lots of stories is an important part of telling the most compelling stories, and the stories that change minds. I think it’s not really structured … Longer would be fine, but longer at the expense of getting lots of ideas out there every week, I think isn’t sort of what Vox has experienced is the best model for what they do and sort of what they know how to make succeed.

Robert Wiblin: Yeah, I guess if you want the really unusual depth then you go and read the citations, read the papers that you’re writing about potentially.

Kelsey Piper: Yeah, absolutely. We do have some long form content. My AI article is more than 5,000 words, which is as long as I wanted it to be. If I wanted 8,000 words, I could’ve done that for that piece, but that didn’t seem like it would add understanding at that point.

How could listeners ameliorate concerns they may have about Future Perfect?

Robert Wiblin: Yeah, is there anything that listeners could do to, I guess, help with the project, or potentially ameliorate any concerns that they have about Future Perfect and its strengths and weaknesses?

Kelsey Piper: I do think people should talk to me. There was a discussion on the EA forum recently where people were airing a lot of sort of confusion about what Vox was doing, and what Future Perfect was doing, and how Future Perfect saw itself as relating to the EA community, and why there were articles on Future Perfect that were about causes that aren’t EA priorities, which is because Future Perfect is also interested in applying some of the same questions and approaches to other topics, and why Future Perfect covers politics when EA has some good reasons to not be involved in politics, which is that Future Perfect is not doing that branding, and is interested in applying some of the same tools and frameworks to political questions. I did get the sense that if people wanted to reach out and say, “Hey, why’s Future Perfect doing this?” Then I could just be like, “Oh, yeah, this is what’s going on.” And people should feel free to do that.

Kelsey Piper: Also, there was somebody was saying we don’t know, for example, whether the Rockefeller Foundation is paying per article, and I would be happy to answer that. Nope, it’s a grant for the year. We don’t know whether they have page view requirements. I’m happy to answer that. No, we don’t have page view requirements, so yeah. I think people can be a little bit more wiling to ask questions if they have concerns, and hopefully at least have concerns that are sort of grounded in a clearer picture of what the incentives are.

How should non-professionals communicate EA ideas?

Robert Wiblin: Yeah, have you kind of got any advice for other people who are trying to communicate EA ideas, like as amateurs rather than professionals? Things that you’ve learned from trying to get people to really understand effective altruism over the years?

Kelsey Piper: I’ve touched on this a bit before, but I think people find effective altruism most compelling when it is answering a question that they have an answer to, and that’s something we sometimes fail at as a community, because for example, I think most effective altruists about like education in the US are kind of like the answer to education in the US is that it’s not a good place to spend money on at this time. But I think when you can, demonstrating how the effective altruist approach gives useful answers on a question that people already care about is a good way to make them care about it as an approach, and therefore care about the answers it gives to other questions.

Kelsey Piper: So without saying anything about what resources should go on the margin, I think it’s often very compelling to say like, “Hey, you’re interested in getting this one particular apartment building at the end of your block approved. You’re interested in homeless shelters in your community.” Or something. Here’s sort of how I would approach that question. I would look at where our money can have the biggest impact. I would look at successful other projects. I would look at what resources seem to be the limiting resources in making this happen. I would do a cost effectiveness estimate. This is the answers I get. This is the guide I would give. And I think when you do that, people are interested in your approach, and are much more likely to care about what answers you get when you ask what’s the best thing to do in the world to compared to it doesn’t seem to have any applications to things that are already important.

Robert Wiblin: Do you get any pushback from people who are like, “Ah, this is so preachy. This is so demanding. This is so moralizing. This is frustrating to me.” I know that’s one thing that EA ideas, one way that they can rub people the wrong way.

Kelsey Piper: Yeah. I think part of the problem there is that people need lots of different things. Like there are people who genuinely benefit from an ethical system that says yes, this is demanding, doing the right thing in the world we live in is actually very hard, and it’s going to ask a lot of you, but it’s important. It will have a lot of returns, and there is supports to help you succeed while you’re doing this very demanding thing. I think there are people to whom that’s really compelling, and then there are people for whom that’s really depressing and overwhelming and they bounce off it, and they feel guilty and scared. And there are people for whom it’s much better to say you don’t have obligations, and it’s not reasonable to ask things of people, but if you could do this little thing, that would make things a lot better. Then there are people for whom that message doesn’t resonate and feels dishonest in some ways.

Kelsey Piper: I think you basically can’t pitch effective altruism to everybody, and I think most people have a comparative advantage at pitching effective altruism to people who are going to find it compelling for reasons sort of like theirs. That doesn’t mean you should just write articles that would’ve convinced a past you, but it does mean that if you’re religious, you’re going to do better at explaining religion to effective altruists than if somebody who’s not religious tries to do that. If you’re coming at it as an environmentalist who really wants to work on climate change, but has maybe decided that’s not the thing to do, then lean into that and talk with other people who care about climate change about effective altruism and about climate change and stuff like that. If you’re here from the animal community, then think about what effective altruism can bring the animal community, and try and bring that as many places as you can.

Robert Wiblin: Yeah, I think it’s important to kind of bring people on the journey with you sometimes.

Kelsey Piper: Yes.

Robert Wiblin: There could be a temptation to jump to final conclusions that are like 10 steps ahead of what someone currently knows about and just bludgeon them over the head with something that … it would make no sense for them to agree or to understand, because you haven’t actually justified it. There could be many steps in the justification, and also I guess just to be like … to not show the same level of curiosity about their interests that you expect them to have about your views.

Kelsey Piper: Yeah, I think those are definitely all errors I see. These days when I try and talk with people about effective altruism, I think I pretty much just say, “Making the world better is really important to me. I’ve gone a lot of weird places in trying to figure out the answer to the question what are the most important problems today, and I think that if you also get in the habit of asking what are the important problems, you will be able to do more.” But that’s not the right message for everybody. There’s not a right message for everybody.

Ideas to promote high-quality journalism

Robert Wiblin: So, we raised some of these concerns about journalism as a business model, and the incentives aren’t great, like the money’s not there necessarily to do the best work. Do you have any ideas, I suppose? You’ve only been working at Vox a couple months, but do you have any ideas of like are there business models that we could that would fund high quality journalism on a larger scale, and maybe also just get it out to more people, and as much as those articles are being written, but aren’t getting promoted sufficiently?

Kelsey Piper: Yeah, I do think there is a lot of high quality journalism, and I think finding authors who you observe consistently doing high quality journalism, and recommending them to your friends, and I think it would be great if a lot more people had a list. Like a lot of people have a list of blogs that they read. I think similarly it’s often valuable to have a list of specific journalists you read that your friends can then click through and sort of see the perspectives that are informing yours. I’d be excited about people doing more of that.

Kelsey Piper: On a bigger scale, grant funded journalism is somewhat promising. Obviously, Future Perfect is funded by The Rockefeller Foundation, but also there’s been efforts to fund journalism that does more investigation. Some of the looking for corruption in public good work.

Robert Wiblin: ProPublica famously kind of follows that model.

Kelsey Piper: Yeah, exactly. I think that’s a promising model if you think there’s a lot of work that does a lot of public good and can’t necessarily support itself with subscriptions or ad sales.

Robert Wiblin: Yeah, so one approach is to write sophisticated content and then find a way to brand it. That gets more attention, so you get more clicks and you can cover the cost that way. I think there is a lot of promise in using foundations, using charitable donations to fund investigative journalism now that kind of the cross subsidy with the newspapers has kind of broken down, but it’s just the total amount of charitable donations as a fraction of the economy is just isn’t so large, and obviously only like perhaps ever a few percent of all charitable donations are going to go to journalism or that kind of political research. I imagine at the moment it’s a much smaller than that, so even if it grew, it’s never going to be as much money as newspapers used to get through subscriptions. I suppose I have no answer to this. I don’t know what the way out is, but yeah, are there any other options that I guess that you’ve heard people ever talk about?

Kelsey Piper: I think it’s going to be difficult for a while. I don’t think journalism is going to go entirely extinct or anything, but I think it’s going to be a low margin, low barrier to entry people who are trying to do something better are in a difficult position with respect to people who figured out something cheaper industry for a while, and then maybe eventually we get UBI. I think there’s lots of people who if they could afford to pay their rent would delightedly report just for the sake of getting stories out there, but yeah, we’re in utopian territory by then.

Robert Wiblin: Yeah. Well, it’s a shame you haven’t been able to solve the entire issues of journalism [inaudible 01:28:16].

Kelsey Piper: Give me six months at the job.

Robert Wiblin: Yeah, so I think you made an interesting point earlier that even though there is like a lot of bad articles, fairly low quality articles in your hot takes of the day style writing out there, it is just true that there’s far more amazing journalism out there than any of us possibly has the time to read, so maybe I’m thinking about this slightly wrong when I’m thinking how do we get more good journalism. It seems maybe what I want to do is discourage people, somehow stop them from reading the bad stuff, which are the kind of candyfloss journalism that everyone, including me, is very drawn to reading on a day-to-day basis, just because it’s so much easier. Yeah, is there anything that could be done to just, yeah, just discourage that?

Kelsey Piper: Subscribe to good newsletters, subscribe to good aggregators. Maybe take the aggregator apps off your phone that will tend to show you what most people are reading, and it’ll be whatever’s the most exciting story of the moment. I think I know a lot of people who’ve benefited from reading once a week about what happened that week instead of reading every day about what’s happening that day. I think there are definitely ways as an individual to make sure you’re reading bigger picture stuff that you care about instead of day-to-day stuff that you mostly don’t care about.

The case for taking AI seriously

Robert Wiblin: Yeah, so one of the articles that you wrote, which I thought was one of the best you’ve done so far, was that the case for taking AI seriously as a threat to humanity. Do you just want to explain kind of what you did in that article and how it came about?

Kelsey Piper: Yeah, so that one was one that was really satisfying to write and really felt like something I valued Future Perfect having and appreciated sort of being able to write. I got several weeks for it and we did several weeks more revising it. The Visuals team did the great artwork for it.

Kelsey Piper: It basically explains why many people, including many researchers in the field, believe that AI is an existential risk and believe that mitigating work, like figuring out how to design AI safety, should start happening now.

Kelsey Piper: My editor did a lot with this piece. He started by just writing like 10 questions he had about the AI risk thing. That meant it was very grounded in, “You’re a smart person who’s heard that EAs care about AI risk, but that’s all you’ve heard necessarily. You have a bunch of obvious questions like, ‘Well, couldn’t we turn the AI off? Like, can AI really get that smart? What would it even do?'”

Kelsey Piper: I feel like that made it a much easier piece for me to write because I was just, “These are reasonable questions that reasonable people are going to have,” and trying to sort of give an answer.

Robert Wiblin: I guess, yeah, this view that artificial intelligence or like very advanced artificial intelligence could pose a big risk. Regular listeners are going to have heard several episodes about this kind of idea before and be familiar with it.

Robert Wiblin: It’s an idea that’s become, I think, like very widespread and like pretty widely accepted now. It kind of used to be a bit more controversial, a bit more of a contrarian position.

Robert Wiblin: I suppose, part of the goal with this article would have been to like take people who are kind of skeptical, who have heard, maybe, an unconvincing version of it or who, for whatever reason, just like have a prior that this is unlikely to be the case, and kind of walk them through the reasons why you think it is a real issue. What stuff did you do to design the article in such a way that it would be persuasive to people?

Kelsey Piper: I think one thing I tried to do is, an objection I hear a lot is, “This sounds like lot of people in Silicon Valley who decided that Silicon Valley is going to save or destroy the world.” I don’t think that’s a very accurate picture of the AI risk landscape. In-particular both, of course, since early in computing, people working in computing were observing that this seems like something that will happen eventually, even if we couldn’t predict when.

Kelsey Piper: Secondly, lots of the people working on AI risk are researchers in universities, not in Silicon Valley. If you look at polls, actually people in Silicon Valley are more likely to be, AI is not a risk than the general public.

Kelsey Piper: One thing I wanted to do was sort of emphasize all of the perspectives that lead you to, “AI risk is a problem,” and hopefully move the conversation beyond, “People in Silicon Valley think that they’re going to kill us.”

Kelsey Piper: Then I think another thing that I ended up coming to during the course of the piece was feeling like, especially since some people now think advanced AI might be developed with building on existing reinforcement learning and machine learning techniques. That meant that a lot of the conversations that I think are very widely happening now about bias, about algorithm transparency, instead of being a distraction from AI felt like part of the same big picture.

Kelsey Piper: Like right now, we design a reinforcement learning algorithm and it doesn’t do what we expect. This has hilarious results if it’s a game playing AI that hacks the game and gives itself a high score. Troubling results if it’s predicting whether people should be paroled AI, that has racially disparate outcomes in violation of U.S. law. Saying, “All right, well, what if those systems were a lot more powerful and what if the scope of their operations was significantly bigger?” That looks pretty bad, right?

Kelsey Piper: I think to lots of people for whom the argument for hard takeoff is complicated or sketchy, just the simple, “We’re using these techniques. These techniques have some very visible failure modes. Scaling them up will mean scaling up the failure modes.” That was something that I think, Dylan found compelling and hadn’t seen before.

Kelsey Piper: That was something that I think a lot of people have an intuition for, which then maybe makes it easier for them to visit the question of, “Okay, what if AI didn’t come from reinforcement learning? Okay, how quickly could we expect this to happen?” Some of the bigger questions that are important.

Robert Wiblin: Yeah, there is this funny catch 22 that sometimes people will say, “Oh, I am not concerned about AI because the only people who are worried about it are kind of, are people who are part of this like Silicon Valley elite, who were working on the problem and have these like delusions of grandeur about how they’re going to affect the world in a really huge way,” so that’s their objection now.

Robert Wiblin: Then you might say, “Oh, no. Actually like they’re not especially more concerned about it than anyone else,” so it’s actually like the general public who’s worried about it. “Well, now I don’t believe it because now it’s only amateurs. Like uninformed people who aren’t close to the technology who don’t believe it.”

Robert Wiblin: It’s like, if it’s the experts who believe that it’s not reliable because they’re biased, and if it’s the general public, “It’s not reliable. Oh, I’m not going to trust them either because they’re not informed enough.” It kind of … It doesn’t seem like both could be right.

Kelsey Piper: This is something when I was writing up the study that found that people in Silicon Valley are less concerned with AI than the general public, I was sort of thinking about how to balance. Because I think my impression of the actual state of the field is that lots of people have seen troubling things with present AI. Some of those I do think are related to the challenges that we’re going to face in aligning advanced AI, some of them aren’t.

Kelsey Piper: Like people are also concerned about like whether self-driving cars will run down grandma. I’ve actually written about this for Vox too. That’s not the same sort of problem as AI safety. It would be misleading to sort of take advantage of people’s concern about that for AI safety conversations. I do think that transparency and interpretability and machine learning conversations are very related to the ones about advanced AI safety.

Kelsey Piper: Then, among experts, the perspectives I see tend to be either, “This is important, we should be working on it.” Or, “We are really far from AGI. We don’t know enough about what it’s going to look like for working on it to get us anywhere.” I think that’s very different than, “Don’t worry, nothing’s wrong.”

Kelsey Piper: Yeah, I want people to have an accurate picture of the field there, but that involves kind of getting into the weeds about, “Why do some people think that this isn’t that far away? Why do some people think that it’s centuries out? How would we be able to tell what the world will look like 10 or 20 years before AGI?”

Robert Wiblin: Yeah, so I guess one thing that people worry about when promoting concern for risk from artificial intelligence and potentially like just all kinds of other risk from new technologies, is that it could encourage kind of competition to create AI more quickly. It can kind of create and arms race if people will start to think that this technology is going to be more important than they felt before.

Robert Wiblin: I suppose, to be honest, like Vox is a drop in the ocean of like this overall discussion of whether AI is going to give particular organizations or countries comparative advantages over others. I suppose, sometimes you might worry that raising alarm about things could be counterproductive for like that or other related reasons.

Kelsey Piper: Yeah, I think there’s definitely, that’s something that you have in mind when you’re writing a story. I think I try to write about AI in a way that makes it as un-sexy as possible, honestly.

Kelsey Piper: If we don’t know how to cause reinforcement learning systems to reliably do expected behavior and have some checks on unexpected behavior that are perfectly reliable, then when they are more powerful and the space they’re operating in is big, terrible things will happen. Not terrible things that benefit the creators, just things that were outside the space we were looking for it and can be catastrophic.

Kelsey Piper: I think people run into this problem a bit more if your perspective is, “This is the most powerful thing ever going to happen. This is the thing that’s going to end death and colonize the galaxy.” That’s part of why I don’t cover it that way, although I also, in part, don’t cover it that way because that’s not really a story I’m comfortable with writing.

Kelsey Piper: Like I don’t feel very confident in speculating about what advanced systems will let us do necessarily so I tend to focus on the case that if we’re not careful, they won’t benefit anybody, including the creators.

Robert Wiblin: Yeah. When you’re writing an article like that, do you have like, do you spend a lot of time thinking about who exactly is the target audience? Do you have kind of particular people in mind who you try to write for?

Kelsey Piper: For that one, my editor, again, was super helpful with a big list of questions that he wanted answered. I think, in general, yeah, I’m writing for … The advice we got in training at Vox was, “Assume that your audience is smart but knows nothing about the topic.” Like, “Don’t underestimate their intelligence but don’t overestimate their background.”

Kelsey Piper: That generally feels pretty good to me because you need to start from the basics, but you can start from the basics with the assumption that you have a smart, informed audience that’s following along and thinking of good questions as you go. That can make for a good piece.

Robert Wiblin: Your colleague, Dylan, tweeted, I think that, I think effectively you or this article or maybe this whole general discussion, he’d been, I guess, somewhat skeptical that, yeah, AI did really pose a threat. To his credit though, he’d, I think, accurately represented at least part of the case in favor of worrying about it in his articles before, but he overall hadn’t been convinced.

Robert Wiblin: It sounds like you changed his mind. Did you guys just talk about it a whole lot in the office and then eventually, like once you presented the ideas in like full sophistication, he realized, “Oh, actually, yeah, maybe this does make sense”?

Kelsey Piper: I think the piece was part of it. I think he’d already sort of been thinking from the perspective that lots of things deserve some resources, and it’s a question of, “What are the most important priorities for the world right now?”

Kelsey Piper: If you’re from the perspective that, “Yep, AI could happen and serve some resources,” then all that really needs to happen is a clearer picture of what the worry is and why resources are useful now. I think there’s probably never as much distance as some of the public skepticism might have implied.

Kelsey Piper: I think, yeah, having a clear articulation of the argument that didn’t rely too much on some of the more esoteric assumptions. I think hard takeoff scenarios are quite likely, but I think that the case for AI safety work doesn’t rest on them, so it’s good to have some cases out there that don’t necessarily assume that, or make it a central feature of the argument, or sort of treat it as, “If you’re not convinced by hard takeoff, you have nothing to worry about here.”

Robert Wiblin: Yeah. I think I’ve become less convinced about the hard takeoff scenarios, but I think that more just changes like how you’d go about things rather than like whether you’re worried or think that there’s a lot of potential leverage by working on that issue.

Kelsey Piper: Yeah, exactly.

Robert Wiblin: Yeah. I guess Dylan might’ve only changed his mind in like kind of subtle ways, but I guess publicly, it seemed like he had changed his mind quite a bit. I think it can be really hard to say, “Oh, well, I think, actually I’ve changed my mind on this issue, where previously I was a skeptic, and I now kind of buy into it.”

Robert Wiblin: I think it’s like super credible that he was willing to do that. I guess it’s a good sign that he’s like a very intellectually honest and curious and open-minded colleague.

Kelsey Piper: Yeah. I really like working with Dylan. He’s going to watch this and then it’s going to be awkward. He’s a great person, super encouraging and helpful and very much someone who’s doing this because he cares about the stuff we’re reporting on and making the world a better place, which I feel like is a key ingredient for something like Future Perfect. You have to be there because you think figuring out what to do is both hard and important.

Robert Wiblin: Yeah, so to what extent do you think kind of educating the public about global catastrophic risk, because as you’re doing, really helps to reduce them? I know there’s some people who think, “Well, yeah. We could make more voters, general, worried about threats from artificial intelligence or all kinds of other technologies,” but then how does that really cash out into lowering those risks? The kind of causal pathways, isn’t super clear.

Kelsey Piper: I think there’s a couple things. One is if you’re a smart, young person and studying or a smart, established person in your career and you see the argument for AI risk and you think, “That’s what I should be doing, I want to do that.”

Kelsey Piper: Are the people around you going, “Oh, yeah. I heard about that. That’s important. I’m glad you’re working on it,” the way I think they would react if you said, “I’m going to work on making organ donations safer, or the way I’m going to work on reducing crime or like improving outcomes in policy.”

Kelsey Piper: Or are people going to be like, “That weird Silicon Valley, apocalypse cult thing.” That matters. I think to a lot of people, that matters. I think all of this is very hard to quantify, but I think if the general Vox reading, smart, wanting the world better, maybe not super informed about AI in-particular, but broadly sympathetic to efforts to handle technology safely. If they’ve heard the case for AI risk and they have this general sense, “Yes, that’s something some people should be working on,” then I think that’s good.

Kelsey Piper: I don’t know how much gains there are beyond that point. Do we get much from coverage of the differences between slow takeoff and fast takeoff scenarios? Do we get very much from differences in approach between, for example, the team at OpenAI and the team at MIRI? I don’t know.

Kelsey Piper: I’m looking for hooks for those articles, honestly. I’d be excited if it seems like there’s a way of telling those stories that people find compelling. Most of the impact I see is that if people are thinking about working on AI, it seems good for the people around them to be aware, “Yeah. That’s something important that we need some people working on it.”

Robert Wiblin: Yeah, I guess writing more complicated or more technical content might attract like more informed and like smarter readers. That might be one benefit there.

Kelsey Piper: I think it’s definitely worth our while to publish articles that people who are in the field are excited about and feel like teach them things. AI is somewhere where I might be equipped to do that, because at this point, I feel like I have a pretty good understanding of some of the technical concerns that the safety teams at various top organizations are working on.

Kelsey Piper: I think Vox does want most of its content, if not all of its content, to be something that a smart person with no background in the area can sort of get into and follow along. I think we’re always going to be walking that balance where we want articles to be informative, even to people in the field and certainly reading as consistently accurate to people in the field. We still want every one of our readers to sort of be able to follow along.

Robert Wiblin: Yeah. How do you balance the incentives to want to create something that will get a lot of hits right away because it’s, say, topical, versus trying to create an evergreen piece that has good search engine optimization and relevant keywords and might continue getting visitors for years?

Kelsey Piper: I think Vox likes evergreen content. There’s a lot of focus on creating things that’ll last. Also, often you can use a news hook and then have most of the story be something that’s longer-term. Like it’s not just about the news, it’s about the context the news fits into and it’s therefore a piece that has a lot of reread value, and that’s definitely the sort of best outcome.

Kelsey Piper: In general, I think I’ll try, of the three pitches I send to my editor, to have like one be a topical news thing, one or two. Then one or two be a just generic big question that I would like Vox to have an answer covering.

16 Big Predictions About 2019

Robert Wiblin: Another article that you wrote recently with Dylan Matthews that I really liked was another one showing kind of the willingness to be open-minded and like more concrete, and try to just improve how people think about issues in journalism was, 16 Big Predictions About 2019, From Trump’s Impeachment to the Rise of AI. Yeah. Do you want to explain what went on in that article and how it came about?

Kelsey Piper: Yeah. I think a lot of people have expressed that it would be really valuable for more punditry and more policy in general to involve specific numerical predictions about outcomes. That’s just because it’s the best form of accountability. It’s very easy to say, “2019, Trump will be a disappointment.” Or, “In 2019, Trump will get some big victories.” Like predictions like that are almost definitely going to come true.

Kelsey Piper: Being specific, giving numerical values lets people sort of tell whether you’re actually someone worth listening to. Whether when you say something is very likely, you’re right. I think Nate Silver won a lot of credibility by doing this in the specific area of election forecasting. He said how likely each outcome state-by-state and then Senate races and House races was.

Kelsey Piper: He’s very well-calibrated. When he says something is 60% likely to happen, it tends to happen 60% of the time. That just means that when it comes to election forecasts, you can expect that Nate Silver is accurately representing his uncertainty in what’s going on.

Kelsey Piper: I think that gives him a credibility that most topics that are less concrete than that don’t have, and that most pundits in any field don’t have. It would be really cool for it to be more widespread, the practice of putting numbers on things and then revisiting, “Where were you right? Where were you wrong?” That’s what we’re trying to do. I think it’s probably going to be very challenging, because it was our first year of doing it. I expect that we got some stuff quite wrong, because this is the-

Robert Wiblin: Forecasting is hard.

Kelsey Piper: Forecasting is really hard, and practice seems to make a big difference in improving at it. I don’t necessarily expect these predictions to make us come out looking particularly good, but I think it’s still really important to do this if you want to set that standard for journalists in general. If you want people to take your predictions about the future seriously, you need to start saying, “Here’s how seriously you should take me. Here’s how good I am at this.”

Robert Wiblin: Yeah. We have an episode with Philip Tetlock where we go into this whole forecasting issue and the question of like why pundits don’t make concrete forecasts, in more detail. It seems like the key reason is that people don’t want to be called out on being completely wrong.

Robert Wiblin: You were saying something like, “Donald Trump will be a disappointment,” is kind of well, obviously, that would be true, but is it even necessary? Because a disappointment relative to what? That kind of claim is so vague that you would never even know if it was true at the end of the year.

Kelsey Piper: Yeah. I think that’s very fair. It’s a claim where the person who made it will definitely, at the end of the year, be able to make the case that they were right, which is different than it actually being true.

Robert Wiblin: Yeah. Do you want to go through some of the predictions that you made and kind of how you arrived at your confidence level?

Kelsey Piper: Yeah, absolutely. We looked at some politics topics. Who is going to be the Democratic nominee? Whether Trump is even going to stay in office. Whether the U.S. would enter a recession. I made that one. I said an 80% chance that we wouldn’t enter a recession, which was just looking at the odds in any given year of entering a recession. Then adjusting up a little bit because it’s been a while since the last recession. There’s some instability and scary stuff happening.

Kelsey Piper: Then right after I made this prediction, the government shutdown dragged on longer than a government shutdown ever has, which did a lot of damage to the economy, and was looking like it might single handedly cause a recession if it kept up for a while. I was like, “Oh, no. Oh, no. I predicted there wouldn’t be a recession. There can’t be a recession now.”

Kelsey Piper: It was amazing how motivated I felt by the fact I’d made this prediction in public instead of by all of the immense human suffering that would be caused.

Robert Wiblin: Yeah. We got to keep our priorities straight. Yeah. “Millions will suffer, but at least I will have been right.”

Kelsey Piper: My 80% is looking pretty good right now. The January jobs numbers looked great.

Robert Wiblin: Yeah. I think, yeah, I think 80% still looks pretty right. What was another one you did here? I guess an 80% likelihood that U.S. homicides will decline this year relative to last?

Kelsey Piper: Yeah. That was one where my own prediction ended up surprising me. A tactic that Tetlock recommends is reference class forecasting, which is if you don’t have a good answer to a question, you just figure out what the general category of the question is and what the answer would be.

Kelsey Piper: For, “Will homicide rates decline?” I asked, “What if I’d asked this question in 2000? What if I’d asked it in 2001? What if I’d asked it in 2002?” It turns out that if you predicted homicides would decline, you would only be wrong four years out of the last 20 years. Every other year, homicides declined.

Kelsey Piper: There are some reasons to think maybe that trend is flattening out. There are some reasons to think that since 2018 was a rise, maybe we should expect regression to the mean. There’s lots of reasons to not be sure that that’s the right approach. I tend to believe that people step away from the reference class a little too readily.

Kelsey Piper: Like if homicide rates declined all but four years of the last 20, including some years when you might think, “Oh, it’s a recession that’s going drive crime up. Oh, there’s other stuff going on that’s going to drive crime up.” Then I think you should trust the outside view a little bit more, so I ended up standing behind the 80%.

Robert Wiblin: Yeah. Interestingly, crime and the economy are incredibly unrelated. Also, interestingly, the Federal Reserve thinks that the amount of time since the last recession isn’t that related to whether, to the likelihood of having another one in a given year. At least that’s what they said a few years ago. Maybe they’ve changed their mind, but yeah.

Robert Wiblin: I think, yeah, absolutely. Starting with the reference class and like you went pretty close to that, is definitely the right approach in most of these cases.

Kelsey Piper: Yeah. We’ll see whether that approach is born out because I also used it in predicting whether global carbon emissions would rise, and in predicting a couple of other things. Whether more animals would be killed for human consumption.

Kelsey Piper: Then one prediction of mine that I wasn’t very sure of, I think has already happened. DeepMind every couple of years … DeepMind is Google’s sister company. They’re both part of Alphabet that does AI stuff. They released AlphaGo and AlphaZero, which was amazingly good at to two player perfect information games.

Kelsey Piper: Last year, they released AlphaFold, which is big deal in protein folding but didn’t reflect any particular advances in machine learning per se. Then just now, they released a StarCraft playing game, which reflects a big improvement over all previous efforts to win imperfect information, complex, real time strategy games, which just are very hard. OpenAI also works on this.

Kelsey Piper: I was pretty excited to see that from them. Even if nothing else happens, I’m going to consider that something within the benchmark I was trying to outline. I had a very hard time describing that prediction precisely enough, because it’s hard to characterize what of their many regular releases is like a big advance on an interesting problem that we hadn’t seen progress on before.

Kelsey Piper: Then, do you count if it’s a big advance but doesn’t seem to use very novel techniques? How important is that? I think AlphaStar, the StarCraft playing game, is a less of a big step forward then say AlphaGo or AlphaZero were, but it’s still enough to make me say, “Yep. That’s the sort of impressive progress on new domains and new problems that you’d expect to see if AI capabilities are continuing to increase.”

Robert Wiblin: Yeah, so you guessed 50% on that. I guess you break even on that, either way.

Kelsey Piper: Yeah.

Robert Wiblin: You can have a strong horse in the race.

Kelsey Piper: Yeah, I wanted to go for a more uncertain number, but sometimes you think something has a 50% chance of happening, and it’s still worth making the prediction because if your 50% predictions are right half the time, then that’s a good thing to know.

Kelsey Piper: Like Scott Alexander does these predictions and he said that his 50% predictions were only right 30% of the time. Which is also a good thing to know about yourself, that when it feels 50-50, you probably have a favorite even if you don’t know it.

Robert Wiblin: I guess another one was, yeah, “Fully autonomous, self-driving cars will not be commercially available as taxis off-sale,” and 90% likely.

Kelsey Piper: Yeah, I am pretty pessimistic. This was one where I didn’t do reference class forecasting. I’ve just been falling self-driving cars and I’m like, Google is generally agreed to be the leader here. They have the most miles driven. They have the most resources. They don’t have any embarrassing accidents that have already happened.

Kelsey Piper: They still seem to be having a really bad time in Arizona, where there’s no weather and no complications, of coming up with a viable Uber-like service that uses autonomous vehicles. That’s not for reasons that seem like they’ll change this year. Like it’s not a shortage of engineer hours, although you can probably eventually patch some of the problem was by pouring enough engineer hours at them.

Kelsey Piper: It just kind of seems like our current techniques that are seeing all these gains aren’t that helpful for self-driving cars, and so it continues to be a really hard problem.

Robert Wiblin: Yeah. I guess, are you worried that in a year’s time, people will criticize you a whole lot?

Kelsey Piper: I worry about all these predictions. Whenever things happen, I’m like, “Oh, no. What if my predictions are wrong?” I think when I am taking a step back from it, I want people to know how accurate I am as a pundit because I want them to have a better understanding of the world, and that includes an understanding of how accurate I am.

Kelsey Piper: If the results come back in a year, then I’m not very accurate, then people should know that and they should read my articles with that in mind, because that’s part of them having an accurate understanding of the world. Of course, I want to be accurate but not checking is very much not the way to handle that. I do think we’ll get better with time.

Robert Wiblin: Yeah. There can be a temptation to give people a hard time when their forecasts are wrong or they make bets about issues and then they’re wrong. I think in a world where most people aren’t doing that, making a bad forecast and losing a bet is actually like way better than not having done it in the first place, so even the loses are winners, to me, in this situation.

Kelsey Piper: I do think, in general, the response has been very supportive with people saying they’re excited to see this and that they commend it. That definitely helps with keeping it happening.

Robert Wiblin: Yeah. Do you think it’s possible to create kind of a competition between different news outlets that don’t necessarily like one another very much of who can make more accurate forecasts? You could have like Vox versus the National Review journalists. Like each making predictions about what will happen that year, and then at the end, like gloating about winning.

Kelsey Piper: It would take a fair bit of time to set up, but I think it would be great to do some sort of cross-outlet thing like this. I know some other people are interested in it. Maybe if this does well for Vox this year, we’ll have a little bit more leverage to convince everybody else to jump on.

Robert Wiblin: It seems like that could drive a lot of clicks. Even if you’re the losing side, or even if, yeah, your journalists are more wrong. Like the competition itself would be very amusing to people.

Kelsey Piper: Yeah. I would be excited about that story.

Things to improve at Vox?

Robert Wiblin: Maybe this is a difficult question but we’ve spoken a lot about things that are good about Vox. What are the things that, perhaps, you like least about it as a publication?

Kelsey Piper: Man, I guess I have to answer that so it doesn’t just sound to our listeners like I came here to talk about how great everything is.

Kelsey Piper: I really wish that it was possible to have more people looking at pieces and more people doing fact checking. I think when I do it myself there’s some things that I’m going to be good at catching, but there are some things that I’m just coming at it from the wrong perspective. And I think I would write better articles if I had that.

Kelsey Piper: I think politics is absolutely part of what Future Perfect is about, because it’s part of how to do good in the world. And there are reasons for EA to not be about that, but they mostly don’t apply to Future Perfect. But sometimes we write politics articles that make my friends who are moderately familiar with U.S. politics be like, “Man, I feel like I’m being sold a side of a story.” I definitely want to avoid that. I want people to feel like they’re not being sold a side. They’re being sold a complete picture.

Kelsey Piper: I think the age of Trump has been very bad for media in that he continually does things that are very annoying and upsetting and not that important, but important enough that it merits coverage. And that means that if you’re a journalist you spend almost all of your time saturated in like, “The president said this blatantly false thing; the president did this unprecedented and rude thing.” And that just creates this environment of exhaustion with the whole situation, which doesn’t seem conducive to necessarily helping people figure out what the most important stories are.

Kelsey Piper: I think the state of media in general is bad for journalists. I think if you don’t really see how you have a viable career in your field, because it’s not clear your field is going to continue to exist, that’s upsetting for anybody. Maybe we’re better at empathizing when it’s people we agree with, but on all sides of the aisle it’s hard to do a good job of a job that doesn’t have a clear business model, and is subject to bad incentives all around, and is pretty frustrating.

Kelsey Piper: Twitter seems very bad for that in that it exposes you to mostly a lot of angry jerks and I think makes people too easily react against angry jerks.

Taxing billionaires

Kelsey Piper: I wouldn’t say this is necessarily a bad thing about Vox, but it’s certainly a place where I notice my views diverging from my colleagues a fair bit is: there’s a big conversation right now about billionaires and philanthropy. I think several Future Perfect articles have touched on this, and other media articles have covered this as well, asking what is the place of people who are in a position to throw $20 billion at a problem compared to the situation where we have higher taxes, and we tax people and we use that money as a society, we vote about how to use it to solve problems. I think most of the people in Future Perfect lean towards the perspective that that would be much better, if we tax the money and spent it on problems.

Kelsey Piper: This is largely a question of which problems you care about and whether you can imagine Congress ever spending meaningful amounts of money on them. If you care about U.S. poverty or U.S. healthcare, or maybe even to some extent international aid, then maybe the government having more money that could be democratically spent on those priorities is good.

Kelsey Piper: If you think some of the most important problems out there are like factory farming, something where the government’s just never going to do much, or if you think that the government is actively working against your interests … I know a lot of sex workers who just wouldn’t be happy about the government having more resources with which to pursue the policies it thinks are correct because it thinks those policies are terrible. Or if you’re worried about the far future and you don’t think that U.S. government work on AI risk is necessarily that helpful since it could contribute to an arms race in a way that private research doesn’t, then I think you have a lot more reservations about the, “Let’s just do this through taxes” thing, because a lot of the things you think are important won’t happen through taxes. If they don’t happen through private charity, they won’t happen at all.

Kelsey Piper: I think I care a lot about various issues that will never happen through taxes, and that makes me more protective of billionaires, maybe, than the typical member of my team. Not billionaires in particular, I do think if you’re a billionaire you have a moral obligation to give away almost all of your money. But I also think that the world where the billionaires are spending it on their priorities is a world where my priorities have a lot more resources than if it’s all taxes and it’s Congress deciding where the money goes, because they’re just never going to care about many things that I think are important.

Robert Wiblin: I’ve read a lot of articles about that, and been thinking about it, and have made some commentary about it on Twitter. I don’t know where I land ultimately. Wouldn’t it be great to transfer the money away from the billionaires and spend it through the government? I haven’t read any article that I feel nails that topic, that deals with it the way that I would, looking at what exactly … What do billionaires actually spend the money on? What would the government actually spend the money on if it raised taxes? What is the actual switch here?

Robert Wiblin: People seem to have very strong views without having a very strong empirical grounding or knowing exactly what the ultimate change would be. And because that’s so hard to know I think there’s a temptation to make the decision for other reasons, like based on what seems just or what seems fair, that kind of thing, which I suppose does matter. But I think it matters less to me than it perhaps does to other people.

Kelsey Piper: I’m certainly not willing to lose Gates’s work on malaria which has saved millions of lives for a fairer distribution that doesn’t save millions of lives.

Kelsey Piper: Yeah, I would also be interested in that article. Maybe I can write it. It’s very hard to write. The question of, “What does the government do with marginal money?” is just such a hard question. I keep stabbing at understanding what the government does with money on the margin, and then backing away because it’s deeply unclear. So that makes it hard to write an article about because at some point your article is just like, “Unfortunately we don’t actually know much about the marginal effect of the government having your money. We don’t even know the sign.”

Robert Wiblin: I think it would be interesting for someone to at least lay out what is the equation that should determine that, even if they don’t know what the numbers are in there. I suppose that would be a step forward even if it would lead people to agnosticism. But I suppose agnosticism doesn’t make for great clickbait.

Kelsey Piper: I could try writing it. I could see what the story there is.

Robert Wiblin: There is this funny paradox. I think the people who currently think that the U.S. government is terribly run and has bad priorities and isn’t taking on the most important issues often seem to be in favor of raising taxes more. I think you can make this coherent, but there’s definitely a strange tension there.

Kelsey Piper: I have also noticed that tension and it also seems a little strange to me. If we raise the money from the billionaires and spend it on the border wall, does anybody who supports raising the money from the billionaires feel enthusiastic about that?

Kelsey Piper: But I think fundamentally the forces that seem to be going on here are that inequality is fairly high compared to most of what people remember, and most of what their parents remember, of their lives. That creates a lot of frustration. Right now a lot of that frustration is being channeled at higher minimum wages and higher taxes, like address the inequality.

Kelsey Piper: And if the ways we address the inequality also make the world a better place then that would be a bonus. But I think on a lot of levels this is a response to frustration about inequality, rather than like-

Robert Wiblin: Working backwards from, “Here’s what I think the government would spend it on, therefore we should raise taxes.”

Kelsey Piper: Yeah, exactly.

Robert Wiblin: I’m very sympathetic to that consideration. I think it should be given some weight. But it’s a very complex policy question. It’s understandable if it’s very hard for anyone to really get to grips with it. I’ve thought about it much. I’ve got no idea, really, where I land.

Kelsey Piper: I also feel like, for me, I’m very much biased by … Most of the billionaires I’ve known as friends of friends, or spent any time paying attention to, have been the ones who are adjacent to the effective altruist movement; or, generally speaking, deeply interested in UBI for making the world better for most people here in general; and then spending their own fortunes to do as much as they can on the sort of problems that they’re more equipped to tackle.

Kelsey Piper: If you look, in my mind, at the prototypical billionaire, it’s like Dustin or Cari who I think are some of the most incredible people alive, or it’s Bill Gates who’s saved millions of lives. That’s superhero territory. And I think for a lot of people that’s not the prototypical billionaire at all. But I think that’s a bias I have. I’m thinking of people who are doing a ton of good with their money. I’m not sure I’m willing to give that up. Or at least giving that up is very salient to me as a trade-off in a way I think it isn’t to everybody.

Downsides and upsides of the job?

Robert Wiblin: Yeah, all right yeah. Returning to the personal level, what are the main stresses of the job? What are downsides that people should be aware of?

Kelsey Piper: Hmm. So I used to hate phone calls. I’ve mostly gotten over that. But-

Robert Wiblin: I was terrified of phone calls when I first got an actual job and had to call people. I was like, Yes, I was like, sweating. But I went away after a few days.

Kelsey Piper: I think it would have been good for me to have a class in college or something that made me make 10 phone calls a day. Just get it over with. Learn that actually, you can do that. I think I frequently find it hard to when I’m reading a bunch of research and I have an overall impression of what’s going on. But I’m not that confident in it, figure out like okay what are the next hour of work on this that will like make me more confident that I’m getting the right impression here. Or pointed out to me if I’ve got the wrong impression here. That’s just always very scary because you don’t have a ton of time.

Kelsey Piper: You definitely want to make sure that you have the right impression of a field. Talking to people is very helpful but often, a lot of them have sort of their take on the field. Especially with all the recent understanding that we’re all coming to of how unreliable research can be and how often published studies just aren’t that good. You know, it can be very tricky to try and do a lit review and say, all right, this is my takeaway. How do I accurately represent to people how sure I am of this? I think that’s something I find hard.

Robert Wiblin: I find it quite depressing when people criticize 80,000 Hours’ work online. I guess, especially when it feels like they’ve misunderstood or they’re misrepresenting it. Have you had any experiences of that with people criticizing the articles? You know, this isn’t fair. That’s not what I was saying.

Kelsey Piper: When I wrote about the particle reactor, I basically said this costs a lot. We might have a complete standard model here. We might not find anything aside from measuring some parameters more accurately. The case that CERN is making for it is saying a lot about there’s so many undiscovered mysteries. Which is true, but it’s not clear that we’re going to solve those undiscovered mysteries with the bigger particle collider. Which doesn’t mean we shouldn’t do it. It just means the cost-benefit conversation needs to be a little clearer about what the benefits are. There was an angry response in Slate a couple days later, that said Vox arguing shouldn’t build a bigger particle collider. Because particle colliders never discover anything. I was kind of like, Hey! I think that’s just kind of how it goes, you know?

Robert Wiblin: Yeah, that’s the way of the world. It’s hard to grasp subtlety or it takes a lot of effort, potentially, and it makes it harder to write the response article, if you like, Well, actually, they had a very nuanced position Then it’s harder to be angry about it.

Kelsey Piper: Exactly.

Robert Wiblin: Yes. So, what have been any particular highlights? Are there things that are especially enjoyable? I guess you’re saying being having access to these experts who are willing to talk to you all the time is pretty great.

Kelsey Piper: Yeah, it’s amazing to call people up and just ask about their research or ask about what they’re doing. I feel like I’ve learned a ton about lots of fields, just by having the luxury of spending a day talking to five experts. Then doing a lot of reading and trying to put together an accurate, if limited, picture of something I didn’t know much about before. So that’s amazing.

Kelsey Piper: I think one thing that’s really great about having a platform is that when I see people sharing an inaccurate view or an oversimplified view, I can respond to that and try and get the idea I think is accurate out there. This happened a bunch with people accusing bed nets of being used for fishing. This is something I run into a lot when I mention that several of GiveWell’s top charities handle malaria and that AMF distributes bed nets. As people will say, “Oh, those are used for fishing.”

Kelsey Piper: Nothing about the cost-benefit analysis for AMF is affected in any way by the fact that some people use bed nets for fishing. They check how many of their bed nets are used for sleeping under and preventing malaria. They go off the reduction in malaria mortality in the area or where the nets are distributed. It’s just, it’s not relevant. So it was nice to be able to sort of write up an explanation of that. Now when I encounter that I can have somewhere to point people.

Robert Wiblin: Yeah, that’s incredibly successful meme. I don’t know why the idea the bed nets says fishing. It’s counterintuitive. I think it’s also people love to find reasons why charities are not effective. And I think in part, it’s because it gets them off the hook for donating to it. Maybe also just people like a story about how they think the world’s not as you expect. There’s also this kind of intuitive contrarian thing.

Kelsey Piper: I think it’s both of those. It’s satisfying to, go oh charity doesn’t work as well as you think, Bill Gates. And it’s sort of satisfying to learn something more complicated. I share both of those intuitions but you got to actually view, right? You can’t just be contrarian.

Robert Wiblin: Yeah. Very true. Do you ever have people just refuse to talk to you? Because like, oh I don’t trust you or I don’t trust Vox or I don’t trust journalists. I don’t want to talk to the media. And then you can’t get information out of them?

Kelsey Piper: Yeah, that happens. And then I go ahead with what I can. Sometimes that means, this can’t be a story because we don’t know enough. Sometimes it just means that the stories like slightly less informed by their perspective. I would like people to know more about the tools they have available to make a conversation with the journalists less scary and more productive. You can say at all only talk off the record and I’m usually completely willing to do that. And then at least you can give me some background and correct my misconceptions.

Kelsey Piper: Maybe I can say okay that one thing you’ve just said, I would love to be able to quote you on that. Is that something you’d be willing to be quoted on? And maybe that is something you’d be willing to be quoted on even if you’re nervous about an unstructured 20-minute conversation? But also I know a lot of people have just had bad experiences with talking to someone who seemed sympathetic and then an article went up that they felt really misrepresented them and their perspective. So that’s just how it goes.

Robert Wiblin: So have you ever had anyone say, “I’m willing to talk to you, but only if I can record this whole conversation. So have a record of exactly what I said. And so like if you ever misrepresent me then I can call that out.”

Kelsey Piper: I haven’t heard that one. If people want to do that, that’s fine. If they want access to my recording, that’s also fine. I think an important part of good reporting is that everyone you talk to, even the ones who you end up concluding are not at all right about their perspective, feels like their perspective was articulated correctly.

Kelsey Piper: Vox does not try to do balance or neutrality in particular. Vox tries to avoid both sides-ism, where you interview two people who disagree instead of trying to take a stance on who’s right. But one thing that Ezra emphasized in our initial trainings on this was actually Steelmanning, which is a rationalist community sort of framing for it. Make sure that your opponent’s perspective is represented. And that you understand your opponent’s perspective and like can represent in a strong way, represent a version of it that readers will find compelling. And then you can argue that it’s wrong. But make sure that you found the most compelling and most accurate. One that’s not weak and easy to poke holes in before you poke holes.

Robert Wiblin: I mean, yeah, so I would talk to you, Kelsey. But I think in general, I’m very reluctant to talk to journalists. I just found that by and large most, like many, many, journalists are trying to find the most provocative take, or the most provocative presentation of what you said, in a way that isn’t super faithful.

Robert Wiblin: I’ve tried to kind of try to do the exact opposite with this podcast. I allow guests to like look over the transcript of the interview and remove anything that they think wasn’t like the absolute best portrayal of their views. If they think that they’ve got something wrong or it wasn’t put the right way. So I want to give people the optimal opportunity to present their views in as much length and as much sophistication as they like. It’s very hard to do that in a nod if you’re really trying to turn a profit as an online media outlet.

Kelsey Piper: Yeah. And I also try for some conversations, particularly ones that are difficult or complicated, to make sure at least that I run quotes by people and that they can say, “Nope, that’s not representing what I meant, even if it’s what I said.” I don’t do that for every conversation more if it’s a big or complicated topic where it feels important. I can see why that’s not something that everybody can do. Nor is it necessarily something everybody should aspire to. If you’re talking to a prominent politician, and they say something about their views, then even if they didn’t endorse it, it may still be newsworthy and worth reporting. I do think if your reporting strategy, it’s mostly trying to get those gotchas instead of trying to talk about the views people endorse, then ultimately, that’s gonna hold back a lot of really important conversations.

Making corrections

Robert Wiblin: Has Future Perfect made many mistakes? Any articles that you’re revert publishing, or you had to make corrections too?

Kelsey Piper: I’ve had to make some corrections. It’s terrifying. I think one thing that this job has given me a deep appreciation of is how difficult it is to write something long and substantive and have everything in it be accurate down to the smallest details. The New Yorker apparently has an entire team that calls people up and asks like, “Is there a duck pond visible out of your window? Is there a red paperweight on your desk?” Just to make sure that everything in the notes is completely accurate. Vox doesn’t have those sort of resources. I fact check my own pieces. I catch errors in the fact-checking process sometimes. A couple times, things have gone to press and still required checking. One time we made Jeff Bezos’ social media press relations team very angry, and they went back and forth with our editorial team for three days on corrections that they wanted to appease.

Robert Wiblin: Did you think it made it better?

Kelsey Piper: I think if I were writing that piece from scratch, I could have better represented Bezos, but I think they also wanted a bunch of corrections that were not about most accurately representing Bezos. It was this odd mix of, “Yes, we can and should do a better job of representing your perspective here but also, some of this is not about more accurately representing your perspective here.” I think everybody should try writing up things they’re very confident in and then checking how many of them are true. I think the first time you do that, and you realize how many of the things you had like that you were just background very sure of were just wrong is very powerful for realizing how hard it is to reliably be accurate. I haven’t had to publish corrections yet on anything that I felt was huge or has changed the conclusions of the piece.

Kelsey Piper: I think it’s very hard to be a journalist for years and not do that; both because it’s difficult for journalism to provide the resources to check all of that and because, even if you’re very careful about checking all of that, the world is complicated. There’s often a bunch of nuances out there that are just very hard to catch. That, an audience of millions will quickly catch for you.

Robert Wiblin: I’ll stick up a link to this great article for … this quite funny article from The New Yorker where they got Daniel Radcliffe, the guy who played Harry Potter to be on the fact-checking team for day. He had to go through like check every fact in an article. It’s actually kind of famous that these American, high classy magazines. I think probably have the most rigorous fact checking off like any publication anyway. I think it may be stronger than like a published academic paper’s doing this.

Kelsey Piper: Oh, I think it is. I think it is. I had a conversation with someone the other day who was like, “But that’s not substantive. Okay, they get the duck pond right, but there is likely as anyone else to get the core questions wrong.” When I read The New Yorkers’ coverage of AI, I don’t always agree with their take, but they do seem to have all of the right facts and get them right in a way that I don’t necessarily see any other coverage of AI. I do think that that sort of level of rigor is related to rigor where it really matters, and I think being very rigorous is genuinely something to look for in coverage. I think people who say, “You read the article. If it’s about something you know, and it has some things wrong, then you should assume it gets that much wrong everywhere else.” I think that’s completely true.

Robert Wiblin: I mean, I also find journalism just as a story journalists put out, it should be pretty exasperating but I think it can be a little bit too easy to attribute to malice what’s more easily explained by tight deadlines and the need to publish a lot of stuff really quickly. It’s like almost all of us would make mistakes all the time in such an environment.

Kelsey Piper: I think that’s true. I think it’s only partially an excuse because I think it’s reasonable for people to say, “Well, I’m only going to read outlets that don’t put their journalists under a degree of time pressure that produces obvious mistakes.” I think that’s probably the right call. Since, in practice, lots of people will read outlets that don’t do that. There’s not much incentive for outlets. Often, they find they can’t keep afloat if they try, and add those delays, and add that process, and add that expense. I definitely think it would be a better world if it wasn’t a competitive disadvantage to do that, and it was a baseline expectation, and there were more resources to get stuff right.

Robert Wiblin: Yeah, I agree that something morally bad and maybe even reprehensible going on here. I guess it’s at the systemic level. It’s a business model level. It’s at the incentives that the readers create level, rather than usually at the level of the individual journalists being malicious or a bad person. I mean definitely there are some like very irresponsible journalists who do harm for the negligence because they get the callousness, but I think that that’s not the typical case.

Kelsey Piper: Yeah, I agree with that. I think working in journalism has definitely given me more of an appreciation of just how high the bar for rigor has to be in order to always get stuff right. That doesn’t mean it’s not important to get there, but I think a lot of people imagine that that’s easier or more straightforwardly achievable, unlimited resources than it turns out to be.

Avoiding the pressure to get clicks

Robert Wiblin: At the moment, I guess the Future Perfect project is being funded by the Rockefeller Foundation. Is it possible that in future there will be like more pressure to make money and get clicks, and there will be like more pressure to produce articles and cut corners a bit more within Future Perfect? Are you hopeful that you can avoid?

Kelsey Piper: The grant is for a year, and then I think we’ll revisit with the Rockefeller Foundation about whether Future Perfect is meeting their goals. I think even within the rest of Vox, it is very highly valued to write good stories that people feel like teach them new things. I think Future Perfect has found a good balance where we are getting lots of page views while still writing content that we’re proud of. So I’m pretty optimistic that in the worst case scenario, where there wasn’t funding for Future Perfect, Vox would be able to keep the section going with the articles that you’re seeing on it right now, because they do get a lot of page views. It seems pretty clear that there is a hunger for that.

Kelsey Piper: I think we’re able to fill that niche while still having high standards. I do think, over time, most things are subject to incentives to get more page views, even if it’s not explicitly, tied to your job performance or anything like that, because you want people to read your stuff so that that tension is always going to be there. I do feel like this is a pretty good environment for having good incentives to produce content.


Robert Wiblin: Yeah, I’m really interested to dive down in this analytics issue. Did you spend a lot of time looking at how many people are viewing your articles at any point in time? I mean, I did this when I published an article. I’m very curious to know, to look at the analytics on the site and see who’s reading it, where are they coming from. You know that thing in your ahead.

Kelsey Piper: I should stop.

Robert Wiblin: It’s addictive.

Kelsey Piper: I should really check once a week or something. It’s important, but getting it as moment-to-moment feedback is not any more valuable than waiting until all the numbers are in. Absolutely, we have a thing that shows the number of people on the page at the moment, where they are on the page. So you can see if they’re actually reading the article all the way through, where they came from, what links they click on, it’s very compelling. I think, we’re all in this because you want to tell people things. When it’s out there, and it’s telling people things, it’s very hard not to check in on them and go, “All right. Who’s that? What are they learning?”

Robert Wiblin: How well do you think he can predict which articles are gonna be of interest to people?

Kelsey Piper: Reasonably well, I think some always surprise you. I think there’s a pool of articles about which I can make the prediction, this has a decent chance of going viral. If it goes viral, then 100000 people will see it. Then some of them do, and some of them don’t. There’s another pool of articles where I’m like about 15000 to 20000 people will read this and then that turns out to be accurate. Global poverty stuff doesn’t do very well. This is something that makes me very sad, and it makes my mother very sad. She reads all my articles, and she’s like, “The global poverty stuff is the best, you should do more of that.” I also would love to do more of that. I think it’s a really important topic, but it doesn’t get nearly as many views or as much attention as both the existential risk stuff and sort of the animal stuff and the weird big ideas sort of content.

Robert Wiblin: It’s really interesting. I haven’t had that experience. I found that the stuff that’s hardest to get to take off for us I think has been things to do with nuclear war and things to do with pandemics. I’ve just never really been able to find an angle on that. I’ve assumed that it’s because people find it so depressing that it seems like it’s easy to come up with a story with global poverty or animals where it’s like multiple, where it’s like more clear what the lesson is and what people want to do.

Kelsey Piper: Yeah, we’ve had some success with nuclear war and pandemic stories. One of my colleagues wrote a story called Here’s How A Nuclear War Will Kill You, which did amazingly well, tons and tons of views. I think people can, if you catch you in the right mood, be down for some.

Robert Wiblin: [crosstalk 00:39:21] So a good article gets 100000 views or so within the first week.

Kelsey Piper: That’s a very good article. I’m very happy about 100000 views.

Robert Wiblin: The average is more like 20000?

Kelsey Piper: I don’t know the average. I think the median is like 15000-20000.

Robert Wiblin: Have you had any stories bomb? I guess you [crosstalk 00:39:52] think these global poverty stories, maybe they just don’t take off.

Kelsey Piper: Some stories, it’ll just 2000 people will see them. In particular, write ups of global development studies will often do that. Sometimes I’m okay with that, sometimes this global development study merited a write-up and a couple thousand people will read it. That’s fine. It still feels to me like content we should have on the site, but it is definitely on the margin when I’m thinking do I write up this global development story? Or do I write up this big idea story? I think knowing what the readership does definitely pushes me in the direction of.

Robert Wiblin: Are you able to come up with theories for why it is that some kinds of stories are successful and some aren’t? What would be the reason why good poverty stories are not have greater interest to people?

Kelsey Piper: I think if they have an obvious simple takeaway, that’s one thing, but often it’s more like this study points at a promising new intervention will need significantly more research to be confident in it but there are some other research that backs up its basic story about its effect. The effect sizes are like consistent with the only other research in the area and like that’s very hard to make compelling. It’s really important. It’s how progress happens. It’s going to potentially save a lot of lives are change a lot of lives. I think reporting on research always runs into the difficulty of conveying how much of an advanced in our state of knowledge is this. Most good research is a nice steady incremental advance in our state of knowledge. It fits in with some existing things. It advances our understanding somewhat. When a study is a blockbuster that completely changes our understanding, then it’s pretty often wrong. I try not to cover something if I’m reading. It makes me go, “What? Really? Wow.”

Kelsey Piper: I think maybe you could get more attention if you’re covering those, but I think also then you would be wrong. I also have something of an aversion. Maybe this is irrational. Sometimes I get pitched stories that I think would do quite well, like, “Hey, we trained dogs to sniff kids and tell if they have malaria.” There’s a bunch of cute pictures of dogs sniffing kids and telling if they have malaria. That would probably do well, but I want Future Perfect to be more focused on the interventions that I expect are going to be a big part of the solution. I don’t see dogs sniffing for malaria as likely to be a big part of the solution. Maybe I’ll be wrong. Maybe it turns out to be a very low-cost way for clinics to test who has malaria.

Kelsey Piper: Then I’ll feel very silly for passing up the story. Where I’m coming from is, if it doesn’t seem like it could potentially be a big part of transforming the global health landscape, then I’m a little reluctant to use those hooks. Maybe I shouldn’t be. Maybe if I want people to read about global poverty, I start with these cool hooks and then I tell them about malaria. We’ll still figuring out that balance.

Robert Wiblin: Yeah, they’re sniffer animals for tuberculosis where I think it actually might seem like it is quite useful because it’s not so easy to diagnose, especially in remote areas. I guess with malaria, it seems like it’s maybe easier just to treat everyone or to use the bed net approach rather than figuring out who has the condition.

Kelsey Piper: Yes, certainly give us current top charities are treatment across the board or prevention. Maybe if they had enough dogs trained, that would actually be an improvement.

Robert Wiblin: I think the difference there is that the treatment for tuberculosis is I think involves months of taking antibiotics. Yeah, it would be like just too much to give whatever. Also, I think most people don’t have it in most places.

Kelsey Piper: We’re of malaria baseline rates are high enough, and the treatment less onerous, so it makes more sense to just do coverage across the board.

Robert Wiblin: Yeah, I’ll stick up the link to that, the sniffer thing. I found it heartwarming and the fascinating. I guess I’m part of the problem here.

Kelsey Piper: No, I think it’s condescending to not tell people stories that they’re going to find interesting and compelling. I do struggle with how do I figure out, but is this important.

Robert Wiblin: You’re talking about how studies that is surprising and interesting more likely to be wrong? I have this theory. Yeah, this one of the kind of society becomes like more reasonable when things that are more likely to be true, are more likely to be copied and promoted and repeated to other people. In as much as things that are more likely to be false are actually more likely to be repeated. We basically like collectively lose our minds, and there does seem to be this kind of effect on social media that’s been created, where it may just actually be the case that the things that are less likely to be true, at least like within many domains, are more likely be repeated. In fact, we just become less than less informed the more we read, at least things that are selected through that filter. I do wonder where it’s like potentially one of the biggest problems in the world.

Kelsey Piper: Yeah, I do worry about how to make sure that true stories are a bigger part of that than false stories, both because it undermines people’s trust and people’s expectation that they can figure out the truth and people’s belief that other people who figured out the truth and published a good explanation are publishing something that they can evaluate for themselves and go, “Yep, this is true. I learned something about what I can do.” It looks very hard to solve, particularly in public research. That’s a community that has been trying to have these high standards for accuracy that realized their standards for accuracy just weren’t working and is working very hard now to figure out new ones. I’m somewhat optimistic about that project. In the meantime, when a stud comes out, I can’t be like, “This has already been vetted. I kind of have to be well.”

Robert Wiblin: I think it can be quite as bad as I was suggesting that because if it was actually the case that last one more likely to be repeated or less equal, then the world will just be like a lot crazier. We wouldn’t even be able to like build houses or accomplished basic goals as a civilization. It’s everything that we do to have some processes for repeating good information in some areas, but I guess within like particular subsets of information, things can be people actually just they’re getting more misinformed over time.

Kelsey Piper: Yep. Anti-vaxx seems like a good example of that. Just a space that is complicated enough that the people who are wrong can come up with evidence as compelling to a lay person is the people who are right. Then you get widespread public health disasters.

Robert Wiblin: I guess it feels like the fight of attention is that perhaps people would all say, “Who would like to know what’s true? They don’t wanna imagine themselves as being someone who promotes things that are false. Truth has that going for it. Unfortunately, it has a downside that what’s true is quite limited to what is actually true. That’s only one thing that’s true, whereas things that are false. It’s like a much broader range of false claims that you can make. You can choose from a much wider range of claims that some of which might be extremely viral or very appealing to people to repeat. You’ve got this constant tension between the preference of people have been saying what’s true and that that capacity to kind of observe and figure it out versus the fact that is just like so many false claims that someone can make. Some of them are just like extremely viral by nature.

Kelsey Piper: That seems to me you see some examples of misinformation media, where it seems like that’s what happened. Lots of people had boring true stories, and then one person had a extremely compelling story that ends up looking like it wasn’t true. The extremely compelling story is the one that gets the coverage because of how outrageous and astounding it is.

Robert Wiblin: Do you ever think when you’re writing a more boring story about an article in some global health social science, you think, “Well, probably only a few thousand people are gonna read this. One of them may will be someone at the World Bank and someone who like actually has access to resources that they have decisions that they can make on this basis.”? Even if just one person in that position reads this article, it justifies the whole thing. Even if it was quite, there wasn’t much reach.

Kelsey Piper: I don’t know that I’ve thought about that in particular, but I have thought that it’s very important to me that the scientists who talked with me about their research feel like the article is a compelling and accurate presentation of what they learned. If that’s true, and if, over time, Future Perfect is something where people doing important research expect that they can get their important research covered truthfully and compellingly, then that’s something I would feel really good about regardless of the views.

Robert Wiblin: Have you learned … I guess it’s early days, but have you learned to use lessons about how to frame different issues in order to make them more interesting to people?

Kelsey Piper: I think I’ve gotten a little bit better at figuring out what kinds of stories the audience is going to find interesting and what kinds of hooks are going to get the audience into those stories, but it’s honestly still a weak point of mine. My editor pretty frequently will be like, “This is a great article that needs a completely different intro that answers these questions that the readers are going to have and is significantly more exciting.” I think I have a tendency to start by answering the questions that Kelsey, at the start of the day had, from the perspective of Kelsey at the end of the day. I have to remember that lots of people are not starting from that point, and need me to back off a little bit and are automatically going to go, “Whoa, new development econ study, sign me up.” You figure out how to make the case, “This is important.”

How did Kelsey build the skills necessary to work at Vox?

Robert Wiblin: Let’s wind back a little bit and find out about what prepared you to be able to do this job, because I think it’s something that relatively few people would actually be capable of doing, producing so much writing every week and managing to make it largely insightful and accurate. How did you build the skills necessary to work at Vox?

Kelsey Piper: I’ve been blogging for a long time, which Dylan tells me it’s a pretty common background for people who end up in things like this, and I mostly write on Tumblr. That’s very different in topic. A lot of it is about personal topics, about disability, about feminism, and LGBT politics. It’s Tumblr. That’s …

Robert Wiblin: That’s its thing.

Kelsey Piper: Yeah, but I think it did teach me how to identify in an argument when there was a perspective that wasn’t being aired, that it would be valuable to get out there, and how to quickly formulate your ideas in a way that people wanted to share and wanted to read. It’s good practice very much so for that.

Robert Wiblin: How long have you been doing that?

Kelsey Piper: I think five years.

Robert Wiblin: Was this is kind of compulsion that you had to share your thoughts?

Kelsey Piper: It was actually a project of Stanford Effective Altruism, which I ran in college. We were sort of talking about outreach and talking about ways of expressing EA ideas. Some of us came up with the idea of starting EA Tumblrs to sort of talk about that. I think I found it a much more exciting platform than everybody else did so I stuck with it longer, but I definitely am indebted to them for getting started in the first place. It has actually been a great outreach tool for EA. I think I talked about what draws me to EA and sort of what I think it’s taught me. For a lot of people, that’s compelling and gets them to look up more and learn.

Robert Wiblin: It sounds like the topics changed a bit from effective altruism to other things that were more the classic topics on Tumblr.

Kelsey Piper: I think, yeah. Maybe pretty early on, it just became obvious that there wasn’t a lot of value in preaching to people on a topic that they weren’t necessarily there for, and that I had a lot of thoughts on the conversations people were already having. Then I think one thing you can do to share any reasoning system, but it works particularly well for effective altruism is just to apply it consistently, in a principled way, to problems that people care about. Then, they’ll see whether your tools look like useful tools. If they do, then they’ll be interested in learning more about that. I think my ideal effective altruist movement, and obviously this trade off against lots of other things and I don’t know that we can be doing more of it on the margin. My ideal effective altruist movement had insightful nuanced, productive, takes on lots and lots of other things so that people could be like, “Oh, I see how effective altruists have tools for answering questions. “I want the people who have tools for answering questions to teach me about those tools. I want to know what they think the most important questions are. I want to sort of learn about their approach.

Robert Wiblin: Do you think you started blogging because you’re already a good writer? Or is it more you became a great writer because you’re blogging for many years?

Kelsey Piper: I think I was already an unusually good writer. I think I was probably an unusually good writer when I was 10. I feel very brag-y saying this. If people are trying to get a sense of should I be a journalist, I don’t know that putting in tons and tons of time writing will necessarily get you there if it’s not something that you already think of as a strength. Although, I do think you have to put in that time to get rid of good at it. There’s a lot of returns to additional practice even for people who are already strong writers.

Robert Wiblin: Interesting, So some people do get better through practice. But if they’re not already decent to start with then, probably they shouldn’t design or they shouldn’t aim for job, where that the core skill.

Kelsey Piper: Yeah, think it is not like programming, and that I know lots of people who had no background in programming and we’re like, “Oh, I think I should do programming.” and taught themselves it at a professional level. I think most people who are writing professionally to had the innate aptitude for writing.

Robert Wiblin: I’m pretty enthusiastic about blogging as a way of putting career capital and perhaps on bias because it’s kind of what I did. I wrote quite a lot of articles when I was an undergraduate, and I think it made me a better communicator. Also, made it possible for you to see that I of knew what I was talking about and like and could think through issues, complicated issues. It’s a way proving yourself and building an audience.

Kelsey Piper: Yeah, I think blogging is a great tool for all of that. I think it’s somewhat meritocratic, in that you can really start up on Tumblr with nobody following you because of who you are, and then develop a following by making good arguments and saying things that people find compelling and having interesting thoughts. I think that sharing your thoughts in public is very good for improving them. I think people will point out nuances, people will point out when you’re wrong. It makes you a lot more thoughtful. I do think there’s lots of reasons to write, regardless of whether you’re pursuing a career as a writer.

Robert Wiblin: Yeah, putting ideas out there was a great way to realize that even when you’re really confident about your idea, the probability that someone is gonna be able come back with a really strong rebuttal is actually surprisingly high.

Kelsey Piper: Yep, that’s definitely something I get out of hearing things publicly.

Robert Wiblin: You said you didn’t find about how to build up an audience, even though I guess most people hadn’t heard of you. How does that happen?

Kelsey Piper: On Tumblr, people reblog. On Twitter, they retweet and you just steadily accumulate followers.

Robert Wiblin: Is it hard to stick with it to begin with, when you’ve got an audience of four friends?

Kelsey Piper: I didn’t find it particularly hard. I think some people like, “I can do that,” but I think especially now, if you start blogging in the EA community, there’s lots of places where you can cross post. There’s lots of places to get reactions to your ideas. Don’t maybe expect tons of reactions right away, but I think if you’re coming up with interesting ideas, you’ve probably be able to get reactions early before you give up on, because it seems like you’re yelling into the void.

Robert Wiblin: Are there any upsides and downsides that people should be aware of about blogging, if they’re considering this might be like something I really wanna invest time in?

Kelsey Piper: I think lots of people are really interested when they get into the EA movement and which bullets they can bite in, which like clever arguments they can come up with. I think if you’re doing outreach, that’s a pretty bad approach because I don’t think that the EA communities like comparative advantages, our ability to bite bullets. I think it’s our ability to answer important questions. I think if you’re doing anything outreach-oriented, if you’re doing it for the sake of like enhancing your own ability to explore ideas, I don’t wanna get in the boat in the way of people enhancing their ability to explore ideas that’s important to just be able to do without PR, thoughts. If you’re doing outreach, I think it’s important to do outreach by demonstrating that you’re good at thinking about important questions, and that your answers to those questions are like valuable and carefully thought through and give people a toolbox to think about those questions themselves. That’s my big outreach advice.

Robert Wiblin: Even faced incentive to be a bit more contrarian or a bit more like, “I have a really striking message early on in order to boot an audience from nothing,” I think I found that when I started writing online because no one has any idea who you are. There’s a certain temptation to be more sensationalist than that’s later on when you already have a have an audience that’s interested to read your articles.

Kelsey Piper: I think I was more contrarian when I got started. Now, I’m thinking was that on some level incentivized by trying to get more readers. What it felt like from the inside was that, when I started everybody on Tumblr was very frustrating in wrong. I had to like explain why with all my clever arguments. Then, over time, I got a more nuanced perspective and a better understanding of what people’s belief systems were doing for them, I guess. Now I don’t find it nearly as tempting to be edgy in those ways, but it’s totally possible that part of the driver of that is wanting to make more of a splash when that’s your way to get readers, and then wanting to alienate fewer people. That’s your way to get readers.

Robert Wiblin: How did it fit in with the rest of your life? It seemed like you would blog a lot on Tumblr. You were writing 1000 words or something on a typical day. Did that like take time away from everything else?

Kelsey Piper: I am a pretty fast writer, so less than you might think. Also, I was practically failing out of college at that time. I was having a very bad time in college. I think, to some extent, I was definitely leaning on the place where I had interesting ideas that people value instead of the place where I was really struggling to keep up.

Robert Wiblin: It’s interesting. Maybe you’re actually building more useful skills through the blogging rather than studying for courses.

Kelsey Piper: I think I definitely recommend to people that they, if they are unhappy in college, and have others stuff they like doing, that they stop the college and do things they actually like. But on the other hand, I’m very lucky in a lot of ways that made that work out for me, and yeah, it’s trickier for people who really do need a degree to make any progress in the area they care about.

Other skills

Robert Wiblin: Yeah, are there any other things that you’ve learned in the past other than blogging that you think prepared you for the role that you’re in now?

Kelsey Piper: I think running Stanford EA was also pretty valuable for that, because it involved lots digging into EA ideas with a group of people. We would have three hour meetings where we would sort of pick a topic and just try to have a much better understanding at the end than we had at the start. I think that was a very valuable experience. I try and get other student groups to do this. They’re like, “Three hours? Nobody’s willing to put in that time.” We got lucky that we started freshman year when we all had a lot of free time, and it sort of was obvious that it was valuable by the time it was more established, but I think the ability to have a question and then just toss around a lot of ideas on it and make some progress pretty quickly is also something that’s pretty valuable in journalism, and just valuable in general. I think everybody in Stanford EA is doing cool stuff now. It seems like in general the experience of tossing around ideas and coming to a better understanding, and just believing that you can do that.

Kelsey Piper: I’m a big believer that if people have the correct expectation that if they think about a hard problem for a couple hours, they will walk away with a better understanding than when they started. That’s just very empowering, and it makes you better at thinking about lots of things, and if you just have the expectation that if you look at something for a couple hours, you’re going to be as lost and confused as before, then that’s really discouraging, and you don’t end up looking into very much, and when you encounter an intriguing or counterintuitive argument, you’re much more likely to be like, “I don’t know how to evaluate this. It doesn’t matter if this is right, because I wouldn’t be able to tell.”

Robert Wiblin: Yes, so you think people underestimate their ability to evaluate arguments?

Kelsey Piper: I think people underestimate it, and I think they don’t learn it, so they correctly estimate that it’s not a strength of theirs, but they don’t see how to improve on that. And I think that there’s a lot of cultural difficulty in … we don’t have a lot of explanations of how to tell whether you can trust your own arguments. We don’t have a lot of tools for people. We have calibration for the specific skill of estimating your own confidence in things, but I can’t think of a good tool that I point someone to if they are like, “I just want to get better at the practice of reading about something for two hours, and then understanding it.” And I wish we had those.

Career advice

Robert Wiblin: Let’s turn now to some more concrete advice for listeners who are thinking, “Maybe I want to be a journalist,” or “Maybe I want to work at Vox,” or some other organization trying to do good by promoting important ideas, and doing investigative journalism as well in the future. Who would you recommend go into journalism, if anyone?

Kelsey Piper: I think that going to journalism, you need very strong writing skills and you need to be a fast writer. I’m not a spectacularly organized person, but a big part of your job will be organizing phone calls and following up on leads, and having phone conversations and transcribing those conversations. That needs to be something you’re at least pretty good at.

Kelsey Piper: Ideally you would also have a pretty deep understanding of the area you want to cover. I think that’s something that can maybe set you apart from lots of journalism candidates, if you’re coming to journalism with a very strong background in something else that’s what you want to cover. But obviously that can be hard for people just out of school who are thinking about what to do.

Kelsey Piper: And then I think journalism is a more social job than I realized. A lot of what you’re doing is you’re trying to get people to tell you about what they’re doing in a way that helps you identify which the most interesting stories are, and which stories people are going to want to read. And you’re working very closely with your editor and with your team to allocate all the stories that need telling and make sure that your stories are in line with your missions and goals. I don’t want to say you should be an extrovert because I’m not and I’m doing okay. But being able to think about people stuff a lot and having pretty strong skills for interpersonal stuff, and having good conversations with people is another thing that’s really valuable.


Robert Wiblin: So you were working as a writer at Triplebyte before. Just quickly describe what that was and whether it helped you on your way very much?

Kelsey Piper: Yeah, Triplebyte does technical interviews with software engineers and then places them at tech companies. My writing work there was very different. I was just writing up each interview and describing what made the candidate talented and what roles they were a good fit for. But it did involve writing lots and lots of content every day, fast, which I think is very much a skill you need for journalism. By the end of my time there I was managing a team, which I think also taught me some stuff about collaborating with other people on writing projects and keeping standards high when you’re doing something at a bit more scale.

Kelsey Piper: I think transitioning from industry like that might work well for a lot of people in that there are more writing jobs in industry. It can give you the chance to hone your skills, write fast, save up enough money that you don’t mind a journalism salary down the road.

Salaries in journalism

Robert Wiblin: So obviously journalism’s very competitive as we’ve been saying. Has that driven down salaries a lot such that it might be hard to make ends meet as a journalist?

Kelsey Piper: I think journalism certainly pays less well than the tech industry, surprising no one. But it’s actually significantly better, at least at Vox, than I would necessarily have expected if you’d asked me to make a prediction in advance. So I make $65,000 a year and that’s with no previous experience. And obviously in the Bay area, that’s not as much as it sounds. But I work remotely. I could work from anywhere if I didn’t have friends out here I bet I would be living quite comfortably somewhere with a little bit of a lower cost of living.

Kelsey Piper: So I think, while obviously it’s not a good career if earning a lot of money is a priority for you, that’s not prohibitive and you don’t have to live in poverty all the time.

How to know if you’re a bad fit for journalism

Robert Wiblin: You mentioned some of the strengths that people need to go into journalism. What are some of the red flags that people might notice about themselves that are just a really bad fit for that kind of industry?

Kelsey Piper: I think if a fast pace at work is something that’s going to burn you out, you probably shouldn’t do journalism. It can be pretty long hours. It can be pretty unpredictable hours as you’re trying to get in a last phone call or something. My editor’s pretty good about telling me, “Don’t work on this on the weekend,” but I still end up working on the weekends a fair bit just to make everything happen. If you get burned out in jobs that have that sort of long, unpredictable hours, then I think it’s probably not a great fit.

How typical is Vox?

Robert Wiblin: Just as a qualifier, do you have a sense of how typical Vox is, or how typical your experience at Vox is, of jobs in journalism in general? Should we just think, “Oh, this is Kelsey’s view,” or do you have a general visibility of …

Kelsey Piper: I don’t think I have that much visibility about the industry as a whole. I think there are probably some jobs that look very different than this. In particular if you’re working for think tanks or something like that you can get some of the benefits of doing journalism but have a much slower expected pace and a lot more time for articles, and maybe a bit more of a 9:00 to 5:00 schedule. So yes, definitely take all of this as what it’s like to work at Vox having transitioned from tech, more than as everything in the field. Although my sister is also working in journalism, so I have her perspective on that too.

Robert Wiblin: We see a lot of potential to do good through journalism but we’ve been a bit reluctant to recommend that people go into it, just for the obvious reason that it’s a shrinking industry it seems. I think the number of journalists has halved or something in the last 10 or 15 years. I was looking at some stats on the Federal Reserve that we could link to and it’s a pretty grim picture. I mean industry’s always changing but it seems like it’s one that’s had a greater share of upheaval and has a more uncertain future than some others. So people should go into it with their eyes open.

How competitive are these roles?

Robert Wiblin: How competitive are these roles? Are there tons of people applying for each of these jobs, and people really shouldn’t count on getting into journalism even if they really are quite a good fit for it?

Kelsey Piper: I think so. I think that, especially if you don’t have your own following from social media to bring to the table or expertise in a highly valued area, I think it can be really hard to get these jobs. There is pretty fiercely competitive application process. I could conceivably see more EA focused journalism roles opening up in the next 10 years that were some of those grant funded or foundation funded journalism with the intent to have an impact, that might change that.

Kelsey Piper: But yeah, it’s competitive. It’s a shrinking industry. Most people with the skills to do well in it would get paid twice as much anywhere else. I like what I’m doing but definitely make sure your listeners know that.

Nearby options

Robert Wiblin: What are nearby things that people can go into that are also good if they don’t manage to get into journalism or media or the press somehow?

Kelsey Piper: If you write articles as a freelancer you’re not going to make a living off of that, but you can do the same thing of getting your ideas on major platforms. And I know more EAs who have been successful with that approach than with being a journalist. There are lots of organizations that’ll pay you to produce content, and some of that might … If you approach it as a journalism role and approach it with the intent to write lots of articles and your organization is on board with that, then you can maybe get a lot of the same benefits with a smaller, but not vastly smaller, platform, and a little more flexibility to make sure you’re covering the things that you think are the most important.

Robert Wiblin: Do you have any opinions on other high impact roles for people whose strength’s in writing and writing quickly, other than maybe being a freelance writer or things that are kind of journalism adjacent?

Kelsey Piper: Seems like lots of direct EA work jobs would benefit from strong writing and analytical skills and the ability to have these conversations. I think a lot of this would translate to evaluating grants for an organization that makes grants, or to working directly at … somebody working on global development or just things like that.

Robert Wiblin: I guess it’s not completely dissimilar to the work that we do. 80,000 Hours, I think, has found it a little bit challenging to find people who both have a really good analytical ability and technical ability, which is something that we need, and also a good ability to write. Someone who can combine those two things can potentially be quite a scarce commodity. I don’t have a sense of, in the marketplace as a whole, how rare are good writing skills. It seems like there’s a lot of people who graduate from the kinds of degrees that give you good writing skills, who don’t seem to have amazing job outcomes relative to people doing technical work. So I’m a bit unsure on the picture here.

Kelsey Piper: I think there are a lot of really good writers out there. I think to do good EA writing you also need a pretty firm grounding in the EA community, so that’s a little more limited. But there are definitely more good writers than jobs for good writers.

Robert Wiblin: One thing is that maybe the people who are getting the most training in writing are doing … They’re writing about liberal arts topics, like literature or something like that. So the thing that ends up being lacking is people who can write about technical topics like economics or machine learning. The people who are learning about that aren’t learning writing, so there’s just not a lot of overlap.

Kelsey Piper: Yeah, that seems like it could be some of it.

What kind of stuff do you wish Vox had the money to do?

Robert Wiblin: What kind of stuff do you wish Vox had the money to do? Do you think Future Perfect could be bigger if it managed to find other foundations that could make grants? Or I guess even individual donors who might be able to provide money to scale it up?

Kelsey Piper: Yeah, I would be pretty excited about that. I think if Future Perfect were bigger we could maybe have some people who are able to specialize in a beat that’s some specific EA cause areas, maybe have some more coverage of factory farming, have some more coverage of global health and development. I think there are definitely a lot more stories to tell than we’re able to get on top of every week, and I would be excited about having the resources to tell more of them.

Kelsey Piper: Also seems like with a bigger team you could maybe do more in-depth reporting on … For example, I always love to report on specific goodwill interventions and things like that. And that’s just hard right now to make work with all of our other priorities. But it seems to me like it would be very valuable for more reporting on: this is a charity, this is what they do, to be available to people who are interested in following those recommendations.

Robert Wiblin: Yeah that’s really interesting. Would it be possible in principle for a listener to give Vox like $100,000, $200,000 and say, “Go away and try to find someone who can write a lot of articles about factory farming?”

Kelsey Piper: I suspect that if a listener wanted to give Vox $100,000 Vox would figure out how to make this happen.

Robert Wiblin: Okay, it’s interesting. I’m not quite sure what I think about the cost effectiveness of that as a donation. If you look at the number of views per dollar that you would get from that it could be potentially good, at least if you’re providing advice that people can actually act on to make the world better.

Kelsey Piper: Yeah, I don’t know if I’m allowed to share details of the Rockefeller grant but doing some quick math in my head, I think you’re certainly buying hundreds of views for a dollar of content that you want to exist by doing something like what the Rockefellers did.

Robert Wiblin: Do you know if other media outlets, other publications, also potentially take donations in order to hire someone to cover a nonprofit issue?

Kelsey Piper: I think if you’re a big dollar donor willing to fund someone’s salary for a year I would expect lots of outlets to have some interest in that. I think you want to be careful about not exerting too much editorial control. Like, “Hey I want you to be able to cover factory farming as a beat,” seems fine, but “Hey I want you to report on how factory farming is evil and bad,” you know, then you’re asking for sponsored content, maybe without clarity to the readers about what is getting paid for. So you’d want it to be something where you want the beat to exist rather than you want a particular angle on coverage.

Kelsey Piper: I don’t know as much about other outlets, but I get the sense that many outlets are interested right now in alternative revenue models. It seems to me like something that somebody who’s interested should certainly reach out and we could talk about what that might look like.

Robert Wiblin: Are there issues around disclosure? It’s interesting. I suppose most people think journalism is funding through advertising, which is not without creating its own pressures for what things get covered and what things get said and what things don’t get said. But I suppose inasmuch as the money’s coming from a foundation or an individual who has an agenda, even if their agenda is improving the world, then I guess maybe you want to declare that on every article, that, “This was funded by Robert Wiblin, who gave money because he cares about this issue, and that’s not necessarily bad but you should know who this person is.”

Kelsey Piper: Yeah, and that’s what Future Perfect has on every article, “Funded by the Rockefeller Foundation.” And then if we mention foundations in the article I’ll tend to also try and say, “In fact, we’re funded by a foundation. They’re paying this salary.”

Kelsey Piper: And I think that’s tough because a lot of people assume there is significantly more editorial control than there is. But obviously transparency and disclosure is absolutely necessary if you want people to have a clear sense of where you’re coming from, and trust that you’re telling the stories you’re telling for the reasons you’ve represented to them.

Robert Wiblin: Yeah, it’s interesting. I guess I haven’t spent a lot of time thinking about this, but I don’t know that I would be more worried about the agenda of the Rockefeller Foundation as any other group. Everyone has an agenda, and advertisers have their agenda. You always want to be aware of it but it doesn’t give me any particular reason to distrust it, the mere fact that it was funded by someone who cares about the topic.

Kelsey Piper: I think the comparison to advertising is sort of instructive there. I hadn’t thought of that. If your alternatives are, “This is funded by a incentive system that rewards as many people reading and sharing this as possible,” versus “This was funded by a foundation that made a grant at the beginning of the year for a full year of coverage that would make the world better,” I actually feel better about the latter when you phrase it that way.

Robert Wiblin: There was this classic criticism on media oligopolies that because they’re reliant on advertising from particular large companies that are selling consumer goods that they would tend to hold back on criticizing those groups for fear that they would punish them by not advertising with that particular outlet. I’m not sure whether that happened a lot but I expect it does. It makes a whole lot of sense. And it doesn’t seem particularly better. Perhaps are you holding back on criticizing the Rockefeller Foundation? Seems like you’d have less reason to do that than to criticize businesses.

General lessons for people dealing with very difficult life circumstances

Robert Wiblin: So you’ve written lots of things on The Unit of Caring, your blog, over the years. I noticed one thing that is kind of related to EA, though it’s not directly about EA, is dealing with mental health issues or dealing with challenges in your life. And I think probably hundreds of people it seems have written in to you with questions about like, “I’m struggling with this or that problem in my life. How do you think I should deal with it?” And you answer with a lot of wisdom. People really respect your opinions on this for that reason.

Robert Wiblin: Are there any general lessons you think you can take away or share from people dealing with very difficult circumstances in their life for someone?

Kelsey Piper: Yeah I think one thing I get out of that is seeing all of the ways that people can be hurting, and all of the ways that the narratives and advice available to them can fail to resonate. And that’s just made me more aware of how many ways you need to communicate something to successfully get it across.

Kelsey Piper: But things that seem to often be important to communicate are that results oriented thinking, that what matters is figuring out how to get outcomes you want and not to beat yourself up over whether you’re having the right thoughts or the right feelings or whether you care for the right reasons, or whether your justifications are valid; trusting yourself more, feeling like in a situation where you can’t believe your own thinking and your own reasoning processes and your own senses. Those are just very destructive situations for people. It’s important to prioritize feeling like your opinions matter, and feeling like you have the ability to reason about reality.

Kelsey Piper: And then just that everybody’s going through a lot. I find work hard sometimes because there are days when I actually just can’t get out of bed and find it very hard to do anything at all. That’s probably a little bit unusual, but the impression I’ve gotten through talking to enough people is that almost everybody has stuff that amount of difficult that they’re dealing with. If you’re imagining, “I’m the one person who deals with stuff like that,” then of course you’re going to despair of having a successful, healthy, fulfilling life. But if you know that everybody else is also dealing with stuff then it can be easier to be like, “Okay, but some of them do have lives that I could aspire to and that I could have.” I think it’s good for people to be aware that there’s lots of really great lives out there that don’t involve fixing everything, but just figuring out how to work around what you’ve got.

Robert Wiblin: My guest today has been Kelsey Piper. Thanks for coming on the podcast, Kelsey.

Kelsey Piper: Thank you so much.

Roundtable discussion

Keiran Harris: Welcome to the “What Did Rob Do Wrong on This Week’s Episode of the Podcast?” podcast. I’m Keiran Harris, joined by Michelle Hutchinson and Rob Wiblin himself.

Michelle Hutchinson: Just like to say how excited I am to be on the Flaws of Rob Wiblin podcast.

Keiran Harris: Oh, we’re so happy to have you, as always.

Robert Wiblin: We’re gonna be stretched for material today, guys.


Keiran Harris: Okay. So, my number one point this week dealt with a section that I wrote about political polarisation. So basically, my principal objection is that we didn’t actually discuss the importance of preventing political polarisation of long terms causes. We mostly discussed animal welfare and global health. I’ve got a few Rob Wiblin quotes here.

Robert Wiblin: Go for it, Keiran.

Keiran Harris: So, Rob said, “It feels like global catastrophic risks just aren’t really that partisan at the moment, or at least in principle, I don’t think there are Republicans who are in favor of nuclear war.” That’s one quote.

Robert Wiblin: Very generous. I like to steel-man the other side.

Keiran Harris: And another is, “I guess I would think it was quite foolish if someone was trying to portray globa, catastrophic risk as Left or a Liberal issue, but I guess I haven’t seen that.”

Keiran Harris: So, my response is yes, there are no Republicans who are openly hoping for a nuclear war, but there are subdivisions of the issue that are partisan, where obviously they shouldn’t be. So, for example, there appears to be a political divide on the question of reducing nuclear stockpiles or eliminating land-based missiles, things like that.

Keiran Harris: So, I found a 2013 YouGov poll that said that on the question of whether the U.S. should unilaterally reduce the number of nuclear weapons, support is 55% among Democrats versus 18% for Republicans. But, presumably, this is just an empirical question. Would reducing stockpiles actually make us safer? Or would it not? Would reducing from, the U.S. has at the moment, what is it, about 5000? Reduce that down to something in the hundreds, the consensus seems to be that we would lose very little in terms of deterrence.

Keiran Harris: And yet, there is this political divide, so when I talk about my concerns about other long-termist issues facing the same fate, I’m thinking along these lines that they wouldn’t be rational. So you and Kelsey often just agreed, saying, “Well, this would be completely irrational for people to disagree on this.” But we see it anyway, and we see a similar thing with climate change, which is how I kind of structured my version of these questions.

Keiran Harris: The idea that we can look at someone’s position on climate change and then, in the U.S., reliably predict their opinions on abortion or gun control, that seems completely insane. But that’s where we are, and so I’m concerned that decades from now we can make a similar prediction based on their opinions around artificial intelligence safety, which, to me, would be kind of a disaster.

Robert Wiblin: I mean, another example is, I think, Trump helped to shut down, or maybe it was Republicans in congress, that helped to shut down the global health security agenda, which was spoke about on the episode with Tom Inglesby about how valuable that is. I think, mostly, because they viewed that as like a foreign aid thing. It’s like something that benefits poor countries rather than something that benefits America, when I guess, in our view, it would do both.

Robert Wiblin:: So, do we all agree that it’s bad for things like nuclear policy to become partisan issues when it doesn’t seem like they divide across?

Robert Wiblin: Yeah. You think it’s likely that they will?

Keiran Harris: No, I think it’s likely that that would be bad, and that we ought to explore whether or not that’s actually likely and look at these previous examples of nuclear stockpile production, climate change, and try and investigate what actually happened there. How did they become partisan issues? Where it’s not obvious that they ever should’ve become partisan, and how we can try and avoid that for questions moving forward.

Keiran Harris: I mean, at the moment, people don’t seem that passionate about it on political lines, so we’re making some progress on, say, lethal autonomous weapons. There isn’t this coalition of people on either side that are arguing against it. You know, people who actually do really care about this can make progress, but assuming that this couldn’t happen, I think is a mistake because you can imagine a story being spun similarly to, let’s say, Cold War thinking of, okay, we need to stay ahead in the arms race. We need to maintain secrecy around our technology, that would be terrible for global coordination for A.I. I can imagine this happening. It hasn’t happened yet, so, you and Kelsey rightfully say, “I haven’t really seen this,” but that doesn’t mean it can’t happen.

Michelle Hutchinson: I wonder how robust we should expect that to be, even if it caused a partisan surge, given that, as pointed out in the podcast, these kinds of issues seem much less partisan in the UK. You might expect that this would be a temporary thing that would last as long as Trump’s presidency but maybe not much longer.

Keiran Harris: Yeah, I wonder though because once you have this perception, particularly in the U.S., it seems to stick. So, the anti-war protests in the -50s and -60s, people who were very against nuclear weapons seemed to have a strong Liberal bias. And then today, I think it still has that Left-leaning association. And we have the same thing for as Rob and Kelsey talked about it in the podcast, for animal welfare in the states. Not necessarily in other countries, but if it was just a one president or a one administration issue, wouldn’t we have seen people returning to the middle on these issues in the United States as well?

Robert Wiblin: Yeah, it seems like when these issues come up, people just try to find the nearest thing that’s a partisan issue and then try to map it on that. So, with the global health security agenda, you’ve got, “Oh, I don’t like foreign aid, so I don’t like this thing,” which involves sending money to Africa. With climate change, you’ve got like pro- versus anti-capitalism, or like pro- versus anti-fossil fuels. With nuclear stuff, I guess people map it onto, like, strong on defense or not strong on defense.

Robert Wiblin: And I guess, for example, with the Ebola issue a couple years ago, I think the Republicans were more in favor of closing off the borders to people from Africa, or some parts of Africa. I think as it happened, that was a pretty stupid policy, but it could’ve been a sensible policy, and I think in that case, Democrats probably would have rejected it on internationalist, globalist grounds. Even if it would’ve made sense. So there’s this instinctive desire to kind of map it onto some, like, existing dispute. So, perhaps that’s kind of the point you’re making.

Michelle Hutchinson: I think the fact that it does seem this slightly arbitrary mapping makes me feel a bit better about the possibility of it being something that you’d be able to swing back if you were trying. Because it makes it feel a bit more like a reason this might not have happened with factory farming is that it just wasn’t that important to that many people, apart from people who are actually raising chickens and trying to make that their livelihood.

Michelle Hutchinson: And, so, you wouldn’t necessarily expect an organic swing backwards, but you might expect that if there was a concerted effort to make a particular issue something that people cared about on both sides of the aisle using the kinds of things that they already cared about, that might be viable.

Robert Wiblin: I suppose that most people assume that it’s very bad for effective activism to be seen as having any political lean. I think that may be unrealistic, and I think that there are some benefits that people don’t really talk about that much that you could get. Where if you do get involved with one political party, then you can potentially have a lot more influence over that party than you would just as an outsider who’s just trying to stay apart from the whole political scene. So, it could be that even though there’s significant downsides to picking sides, it’s very hard to get things done without doing that to some degree.

Michelle Hutchinson: Yeah. I wonder whether there are ways of getting the best of both worlds by having different parts of this linked with different sides so that the overall worrying about existential risks ends up being not polarised because different issues within that are polarized in different ways, so you might think that nuclear disarmament is a Left kind of issue. And then you were saying that biosecurity could even be seen as a issue on the Right because Conservatives will be more willing to act as national groups rather than internationally.

Keiran Harris: Yeah, so I mean I would be excited to see long-termist causes being treated like foreign policy is in newspapers. So, at the moment if people wanna read about A.I. safety, if the only place they can go is Vox, which I think one reasonable critique might be of this episode is that Rob framed Vox as being center-left.

Keiran Harris: I think if you looked at analysis of media outlets you would put the Washington Post center-left. You would put the Atlantic at center-left. I think Vox is beyond that. And, so if they are the only outlet who are talking about these issues, then it would be reasonable to be concerned about a perception of a Liberal bias there.

Robert Wiblin: I guess as Kelsey said it’s probably easier to persuade people who have other political views to go and be outspoken advocates for these issues than it is to convince everyone who happens to be liberal or progressive to just stop talking about animal welfare, or stop talking about existential risks. It’s a lot to ask for someone to gag themselves like that just because they happen to have the most common politics that’s associated with that view.

Robert Wiblin: I guess I think having a first, kind of conservative, outspoken person talking about existential risk could be really quite valuable. So, if someone had that as an option, I’d be surprised if it wasn’t among the top handful of things that they could do.

Michelle Hutchinson: Yeah. I think this discussion has been fairly U.S. centric so far in terms of describing where various news outlets seem to be on the U.S. political spectrum. There’s a pretty exciting new project that’s recently started in the BBC called BBC Future, which seems to be trying to do a somewhat similar kind of thing to Future Perfect, and the BBC doing it, from a UK perspective, seems really good because the BBC is seen as pretty neutral by both Conservatives and Labor in the UK.

Michelle Hutchinson: On the other hand, I think Americans would typically see the BBC as pretty Left-leaning, so it’s not clear how this would stand for Americans.

Keiran Harris: Do we think that, potentially, it’s justified to have a quite U.S. centric view on this, given the outsized influence that the U.S. government might have moving forward in terms of regulating A.I.?

Robert Wiblin: Yeah, I guess the U.S. is gonna be the dominant player, or China, which I guess is a little bit outside our area. I suppose the UK could have an influence. Like, DeepMind is based there. I think the UK, at least in the past, had some influence over the EU. I guess we’ll see where that ends up going forward. But, I guess if you could convince bureaucrats in the UK, the might be able to pass the message on to the U.S. There’s like, at least some potential influence there, and likewise in Australia, but perhaps a bit tenuous.

Robert Wiblin: I guess we’ve ended up talking about the U.S. mostly because Vox is a U.S. based organization that, I guess, has most of its audience here and covers American politics a great deal. Perhaps, also, it seems like there’s a greater risk of effective activism being viewed as on one side here. Because in other countries, I think throughout, people who are involved in X-risk work who exist across the political spectrum more than you get in the U.S. at the moment.

Specialists in journalism

Michelle Hutchinson: I’d be interested in chatting a bit about what we think the implications of some of the things you guys discussed should be for EA’s, specifically in their career decisions. One thing that I was thinking about, listening to it, was I had previously thought that being a journalist was the kind of thing that you probably want to go into early in your career and you want to be a kind of generalist in terms of what different things that you write about.

Michelle Hutchinson: But Kelsey mentioned the possibility of having a specialist on, say, A.I. and that seems to be a thing that we haven’t talked that much about, the idea that, people partway into their career, having done a PhD and maybe worked in policy for a while might then want to go into journalism and work somewhere like Vox in a very specialist area, producing really high-quality information on that area. What do you guys think? Is that a promising pathway?

Keiran Harris: This seems like it would tie into the project of creating plausible platforms for this work. So, Kelsey talked about, you wouldn’t wanna just create an A.I. vertical anywhere. You’d wanna have the right people there, presumably if we had people who were experts in a particular domain. They wouldn’t necessarily wanna make a career change unless there was a place for them to work.

Keiran Harris: So, potentially we would want to have people setting up these verticals at respected institutions first and actually create these jobs. And then, we could look to move people into them. But I wonder how plausible it would be for someone today to make that shift, or commit to making that shift, knowing that the only actual place that’s doing this is Vox and Future Perfect.

Michelle Hutchinson: I guess the kind of thing I was thinking of was mostly people keeping in mind that this could be the kind of thing that they would want to do. And that would affect their immediate actions in terms of maybe they would choose to do more writing, maybe they would choose to do more writing, maybe they would choose to have their own blog, maybe in the style of Owen Barder or something like that, but in different areas. But it wouldn’t be as serious as actually looking for jobs immediately, and that way perhaps it could be a more organic thing where they would at some point discuss this as a possibility. And so the vertical would be set up with a particular person in mind who people already knew was interested in this.

Robert Wiblin: Yeah, it seems like, kind of no matter what expertise you develop, this is a potential option later into your career is to go into advocacy more generally, like making use of all the things that you’ve learned.

Robert Wiblin: It also reduces some of the downsides of going into journalism, that if you actually know a subject really well, then probably you’re going to be better than the other things that people could be reading about that, which is much less obvious if you just leave college and then just start writing as a journalist. It’s not entirely obvious that you’ll be better than what people would have been reading otherwise, or would be accurate enough to brand effective activism well.

Keiran Harris: Yeah, do we feel comfortable advocating strongly that people start their own blogs? Presumably, that will take away from time that they could’ve been spending on high-impact things in the short term.

Michelle Hutchinson: I think the thing I feel happier advocating is people having a go blogging, where that might mean writing something up for the effect of altruism forum. I would expect that writing up a few posts in a few days would be pretty useful for someone as an idea of do I enjoy writing somewhat fast? Do I enjoy getting clearer on my ideas and that kind of thing?

Michelle Hutchinson: And then it would be a relatively useful exercise for most people because it would get them to think more clearly about some particular issue that they had kind of wanted to be thinking about. Whereas I would be much more hesitant to suggest they actually set up their own blog, as you say, because that would be such a time commitment.

Robert Wiblin: Yeah, Michelle. Was there any other reactions you had to the episode?

News aggregator project

Michelle Hutchinson: One thing I was thinking as I was listening was whether we should be excited about the idea of a news aggregator that was set up deliberately as a charitable project, which was aiming to make someone as well informed as possible in terms of the big picture of the world.

Michelle Hutchinson: I don’t tend to read much news at all because I find it easy to be sucked into things that actually aren’t that relevant, or which aren’t very action-guiding for me and I’m very grateful to a lot of people in the EA movement, particularly Rob for coming up with specific ideas of things that I can do in cases where, for example, going door to door asking people to vote in the Brexit vote.

Michelle Hutchinson: And, I think I would find it pretty useful to have a news aggregator which wasn’t incentivized just for clicks or for anything like that, but was specifically targeted to get people to have a good understanding, both of the world at a high level and of which things we should act on now. And wasn’t just trying to get you informed on things that were particularly inflammatory.

Keiran Harris: That sounds like a really exciting project to me. Do you have a kind of person in mind who would be a good fit for that? Someone who you would be maybe coaching at the moment? If we had that project set up you would say, “Maybe you should consider this?”

Michelle Hutchinson: I expect it would need to be someone who, themselves, enjoyed reading a huge amount and reading things on lots of different news sites. But also someone who is quite good at shifting reference frames because I think a lot of this would be trying to abstract away from the way typical media outlets present things. And think about what’s the thing here that people actually need to know, if anything, and how should I present it so that they don’t get sucked into “this disease is really bad, but maybe, oh, well actually, if you look over a twenty-year time horizon, this disease is really dwindling”. And so, the story really here is this particular vaccine has been surprisingly successful. That kind of frame changing that’s a bit difficult to do.

Robert Wiblin: People have joked at various points over the years about creating a utilitarian news site, where I guess you would try to measure the badness of all of the problems that you’re describing in terms of their actual effect on welfare. I think that could be quite an interesting experiment, though I’m not sure whether it would get enough readership to continue. But I think I would find that very amusing to read.

Any updates in our perception of journalism as a career?

Keiran Harris: Do we feel like the episode gave us enough information to update our position on how strongly we should recommend journalism as a career?

Michelle Hutchinson: I think it gave me more of a sense of the day-to-day of journalism. I thought that she did a really good job of that. And then, she said her sister’s also a journalist, so I think she may have been being relatively modest about how representative the types of things she said were because I would guess that at least she had some sense of when her particular experience was not generalizable and was kind of keeping that in mind.

Michelle Hutchinson: I still feel like I don’t have a super strong sense of how valuable journalism is in general and in particular how much it varies depending on whether you’re on a kind of typical beat and waiting for the one or two useful stories to come up, versus working at Future Perfect or BBC Future.

Robert Wiblin: Yeah, I guess Kelsey’s job seemed more appealing than what I imagine most journalism jobs would be. She has a particularly accommodating team and a lot more editorial freedom than perhaps most journalists would have.

Robert Wiblin: I suppose in as much as more positions like that open up that are funded through donations, it seems like it could be a better career path than what I guessed before. I guess it’s really gonna depend on the specifics of someone’s situation and what role they’re going into.

Robert Wiblin: Like, I’d be very happy for the reasons we discussed earlier to see some Conservatives writing about X-risk in the Wall Street Journal or something like that. I think that would fill a real niche where that audience isn’t really getting that content.

Robert Wiblin: But it just depends on whether that’s actually an option that exists. I think, also, someone who prepares themselves to be able to do what Kelsey’s doing, to like, write quality articles about important issues quickly, is going to have a lot of other options, even if they can’t get into journalism. There are gonna be a lot of movement building, or research roles, I think, that would be in a good position to take.

Michelle Hutchinson: Yeah. I really enjoyed this episode and listening to Kelsey talk about how she felt about these things. I was incredibly impressed by how fast she has output and learned about these issues and how deeply she thinks about them despite the fact that she has to produce five articles a week. And I was really glad to get this insight into her job.

Keiran Harris: Fantastic. Well, thanks for joining me on the What Did Rob do Wrong This Week Podcast.

Michelle Hutchinson: It was my absolute pleasure, Keiran

Robert Wiblin: Always love to have people talking about me.

Robert Wiblin: Just a reminder that applications to come to Effective Altruism Global in San Francisco are now open at eaglobal.org.

And if you enjoyed this episode please take a moment to send it to a friend.

The 80,000 Hours Podcast is produced by Keiran Harris.

Thanks for joining, talk to you in a week or two.

Related episodes

About the show

The 80,000 Hours Podcast features unusually in-depth conversations about the world’s most pressing problems and how you can use your career to solve them. We invite guests pursuing a wide range of career paths - from academics and activists to entrepreneurs and policymakers — to analyse the case for and against working on different issues and which approaches are best for solving them.

The 80,000 Hours Podcast is produced and edited by Keiran Harris. Get in touch with feedback or guest suggestions by emailing [email protected]

What should I listen to first?

We've carefully selected ten episodes we think it could make sense to listen to first, on a separate podcast feed:

Check out 'Effective Altruism: An Introduction'

Subscribe by searching for 80,000 Hours wherever you get podcasts, or click one of the buttons below:

If you're new, see the podcast homepage for ideas on where to start, or browse our full episode archive.