Whatever people think about this particular reply by Nonlinear, I hope it’s clear to most EAs that Ben Pace could have done a much better job fact-checking his allegations against Nonlinear, and in getting their side of the story.
In my comment on Ben Pace’s original post 3 months ago, I argued that EAs & Rationalists are not typically trained as investigative journalists, and we should be very careful when we try to do investigative journalism—an epistemically and ethically very complex and challenging profession, which typically requires years of training and experience—including many experiences of getting taken in by individuals and allegations that seemed credible at first, but that proved, on further investigation, to have been false, exaggerated, incoherent, and/or vengeful.
EAs pride ourselves on our skepticism and our epistemic standards when we’re identifying large-scope, neglected, tractable causes areas to support, and when we’re evaluating different policies and interventions to promote sentient well-being. But those EA skills overlap very little with the kinds of investigative journalism skills required to figure out who’s really telling the truth, in contexts involving disgruntled ex-employees versus their former managers and colleagues.
EA epistemics are well suited to the domains of science and policy. We’re often not as savvy when it comes to interpersonal relationships and human psychology—which is the relevant domain here.
In my opinion, Mr. Pace did a rather poor job of playing the investigative journalism role, insofar as most of the facts and claims and perspectives posted by Kat Woods here were not even included or addressed by Ben Pace.
I think in the future, EAs making serious allegations about particular individuals or organizations should be held to a pretty high standard of doing their due diligence, fact-checking their claims with all relevant parties, showing patience and maturity before publishing their investigations, and expecting that they will be held accountable for any serious errors and omissions that they make.
One area where Ben didn’t follow investigative journalism “best practices” (that I had missed early on, but saw mentioned in Kat’s post, and went back and checked) was that he financially compensated his sources ($5,000 each, or $10,000 total). This is frowned upon pretty heavily in investigative journalism (see e.g., https://en.wikipedia.org/wiki/Chequebook_journalism). I don’t have any reason to believe this meaningfully distorted the outcomes here (for instance, if the sources had no indication right until the end that Ben would compensate them financially, it is unlikely to have influenced their behavior) but it is a clear departure from an existing norm in the investigative journalism field. I appreciate that Ben disclosed this information; disclosure does address some but not all of the concerns around compensating sources.
I don’t rule out the possibility that the investigative journalism norm against paying sources is flawed, or it doesn’t apply in this case, or that a different set of norms should be applied.
The Commission is authorized by Congress to provide monetary awards to eligible individuals who come forward with high-quality original information that leads to a Commission enforcement action in which over $1,000,000 in sanctions is ordered. The range for awards is between 10% and 30% of the money collected.
The linked Wikipedia article also has many quotes saying that really the central problem here is disclosure:
The key issue is disclosure. If you pay for it, say so, so the viewer can draw whatever inferences are appropriate about the veracity of what the paid source is saying.
– Jerry Nachman,former vice president of MSNBC news[27]
Habryka—Note that when the Securities & Exchange Commission (SEC) offers substantial whistleblower bounties, they do not simply take the whistleblowers’ word for it and start arresting people. They apply huge teams of auditors, lawyers, investigators, and federal agents to see if the whistleblower allegations are good enough to warrant legal action. If in doubt, they might convene a grand jury to see if the evidence is strong enough to take to trial. And they know from bitter experience that if they simply offered bounties to anybody who makes allegations, they would be deluged with false accusations.
Likewise with journalists. Yes, they offer payments in some cases for whistleblowers. But good journalists fact-check, with the expectation that many ‘whistleblowers’ will turn out to be bad actors with dubious agendas.
If we incentivize accusations, we’ll get a lot of false accusations. There has to be some good-faith effort to check if they are actually true.
Yep, I agree that it’s ideal for the prizes to be paid out conditional on them passing fact-checking or leading to a conviction. I think that was hard for various reasons in this case, but it seems clearly better for incentives.
I do think there was quite substantial good-faith effort of fact-checking involved. It might still be the case that it failed, I am still reading the giant 135-page document, but as many have pointed out, this investigation involved hundreds of hours of effort and interviews with over a dozen people, dozens of drafts, and many many fact-checks. So I do object to you implying there wasn’t any such process.
I think implying that the process that was present wasn’t enough and there should have been more, or that there were substantial issues with it, seems like a reasonable critique that I am still thinking about, but I think implying the absence of one seem bad.
My key point about investigative journalist expertise is that amateurs can invest a huge amount of time, money, and effort into investigations that are not actually very effective, fair, constructive, or epistemically sound.
As EAs know, charities vary hugely in their cost effectiveness. Investigative activities can also vary hugely in their time-effectiveness.
Yeah, I can totally imagine there are skills here that make someone substantially more effective at this (I think I have gotten vastly better at this skillset over the last 10 years, for example). As I said, I think criticizing the process seems pretty reasonable, I highly doubt that we went about this in the most optimal way.
I’m glad you’re taking time to read through the evidence and think on things.
For fact-checking, a basic thing would have been to speak to us and see our evidence. Like, we had interview transcripts and text messages and lot of relevant things. We reached out to Ben multiple times to talk to him and share this.
Ben updated a lot when he spoke to us. There was a lot of things they had told him that turned out to be false. He had reason to think we had more evidence like this, like what we’d already told him about and shared.
He also had all of the information needed to know that we weren’t the retaliatory villains Alice and Chloe painted us out to be, who would attack them for sharing their side. They had been telling bad thing about us for 1.5 years and none of their fears came true.
The only thing we ever did was share our side, which they tried to portray as unethical.
It’s not that Ben didn’t have a process. It’s that his process would predictably lead to incorrect conclusions. Spending less than 1% of the time talking to and trying to understand the other side and spending 99% of the time on one side is soldier mindset, not scout.
I continue to think that our concerns about retaliation were well-placed. Ben did talk to you multiple times, which I think clears the basic bar for due-diligence.
Sadly due to the track record of retaliation that you do seem to have displayed, I continue to endorse not engaging with you further during the investigation, though maybe I will change my mind on that after reading more of the evidence document.
I really wish things were different and we could have collaboratively investigated the accusations, but man, yeah, the libel threat was really bad, as were a bunch of other things that we heard about you and you said to us directly, and also of course we were concerned about retaliation to our sources and didn’t see a way to avoid exposing them to more risk from you without having things public.
There isn’t a track record of retaliation. We didn’t retaliate against your sources. We know who almost all of them are and what they said and nothing happened to them.
Alice’s messages simply show me saying that if she continued sharing her side, I would share mine. Sharing your side is not unethical.
For the libel, Ben knowingly said multiple things that were false and damaging, and he said dozens of things that he could have easily known were false if he’d just waited a week out of 6 months.
But we never wanted to sue Ben. We just wanted Ben to give us time to look at the evidence we were more than willing to share with him. I really recommend reading this section, because I think it gets across very well what was happening.
Here’s a quick excerpt:
However, saying it’s wrong to threaten a lawsuit with Lightcone would be like if somebody drew a gun on you and you tried to knock the gun out of their hands. If the gun-wielder then goes around saying that “you hit them” they’re missing a critical detail in the story.
Ben knowingly published numerous falsehoods that were extremely damaging to us. He published dozens more libelous falsehoods from Alice and Chloe which he could have easily avoided if he’d just looked at the evidence. He knew he was about to wreck our ability to do good and cause immense personal suffering to us.
He heard somebody—who has a reputation for dishonesty—yell “thief!” and shot us in the stomach before he could check and see if we were actually thieves. He was unwise and reckless. You shouldn’t shoot people that easily. Especially when you know that the person yelling it has told you lies before. But he was well-intentioned nonetheless.
And even as Ben proudly says he’d shoot us again, we’re saying that the real villain is unaligned AI and let’s focus on that. We should not be fighting each other. We don’t want to fight.
Ben, we’re on the same side. We all want to make AI go well.
You were misled. By women who need help and compassion, no doubt. But the way to help them isn’t to shoot us. It’s to actually try to understand the situation, then go from there.
Remember: the way that good people do bad things is to demonize the other. So even if some of you might be very mad at Ben for doing this, I call on everybody to try to be their best selves. To set off an upward spiral. To remember that Ben had good intentions. His methods were bad and the outcomes disastrous, but the way to solve that is not to shoot him. The way to solve it is to creatively problem-solve, assume good intent, and remember the bigger picture.
To always remember:
Almost nobody is evil
Almost everything is broken
Almost everything is fixable
And the accusation of threatening to hire stalkers is just a really weird accusation. That should be an indicator that somebody is not mentally alright.
I’m really sad too that we couldn’t just talk too. I hope we still might be able to, once you’ve read the document and see that the retaliation reputation was unwarranted. I would really love to talk. I think trying to do conflict resolution in a high stakes, hostile, and public venue is less likely to work than if we can talk face-to-face and have a higher bandwidth conversation.
Honestly, I wish we’d already invented mind reading technology, because I’d just let you read my mind, unfiltered. I know that if you could, you’d see that I really have no negative intentions and I’m really just trying to figure out how to make everybody happy and reduce suffering. This situation is complicated and I certainly can sometimes unintentially cause harm, and I hate that, and I’m always working on trying to prevent that. But I really do just want everybody to be happy, including you. Anyways, for now we don’t have mind-reading technology that’s accurate or cheap enough, so we’ll have to make do with me trying to convey through text that I really am not retaliatory. If you hurt me, I will try to understand you, try to help you understand me, then try to collaboratively problem-solve.
My original comment left a pretty wide window of possibilities open, and your reply falls within that window, so I don’t quite think we disagree a lot. However, in the spirit of nitpicking, I’ll make a couple of points:
Prominence of disclosure matters. The fact that Ben included the information in his post shows that he didn’t intend to hide it; nonetheless, my sense is that he didn’t highlight it as a disclosure / disclaimer / caveat for readers to keep in mind when interpreting the post. He did include other disclaimers around his process and motivation at the start of the post, that I found helpful, and his non-inclusion of payment along with those disclosures gives me the sense that he didn’t consider the distortionary effect of payment as a biasing factor worth highlighting to readers. My guess is that it would be pretty likely for readers to miss it (as I did). I’m genuinely uncertain whether the lack of discussion around this was driven by people not noticing it, or noticing it and not thinking it mattered.
I’m familiar with the broad outlines of the whistleblower law (from this podcast episode). I think there’s a distinction, though, between awarding money after a determination / judgment of harm, versus awarding money as a journalist or investigator who’s trying to report on the situation. I don’t know exactly how Ben perceived his role, and perhaps the point is that he didn’t perceive his role as being strictly one or the other, but a mix.
Yeah, this makes sense. FWIW, my current memory of the situation was that Ben hadn’t made any promises about paying for information until quite late in the process, and the primary purpose of the payment was to enable the publishing of information that was already circling around privately (i.e. in private docs that Alice and Chloe had shared with some others).
Of course, it’s hard to get rid of the incentive, but I think given that it was paying for publishing something that was already largely written up, I do think the immediate incentives here are weaker (though of course in the long run, and also via various more TDT-ish considerations, there is still an effect here).
I also am not super confident in the exact historical details here. Slack records suggest the rewards hadn’t been finalized the week before the post.
Alas, I do not know. I have some internal Slack records suggesting it as a thing to consider in April, but I don’t know when Ben brought it up to Alice or Chloe. I am confident nothing was confirmed until quite close to the post being published, but I don’t know when the idea was first floated (with the only bound I have is that it probably wasn’t before April).
“High-quality information” is key. Ben did not fact check basic things and we’ve provided evidence that a huge amount of the “witnesses testimony” was false or misleading.
If Ben had waited one week out of the 6 months he spent working on this, he would have known this and not trusted the sources. They said dozens of provably false things and we just wanted some time to share it all with him, since they’d accused us of so many things.
We didn’t want ages—we just wanted a week. And Ben had been working on it for 6 months, so it didn’t seem like that much to ask.
Paying people $10,000 to say untrue or misleading information seems bad. People should not be paid until their facts are checked, and if it’s shown that their facts were false, they should not be paid.
Agree that in as much as people were paid directly for propagating inaccurate information, then that seems sad and clearly sets the wrong incentives. I am not currently convinced of that after the initial reading of your evidence document, but I am still reading, and there really is a lot of stuff to process.
In this thread I am trying to have a locally valid discussion on the actual presence of norms against paying for information among investigative journalists. I would prefer if we can keep the discussion here more local since it seems like an interesting and somewhat important question, and I think it would be an important update for me if there were was a consensus among investigative journalists and similar professions that whistleblower prizes and paying for information is a bad idea.
This is also why I would be very wary of the EA Community Safety team offering ‘whistleblower support’ (which could boil down to ‘bounties for false accusations’).
I think it does incentivize them to distort what they say. They were incentivized to make everything sound maximally sad-sounding. Ben said if they did the emotional labor of sharing their sad stories, he’d give them $10,000.
They knew that if their stories hadn’t been very sad (e.g. Alice said she did get food but it just wasn’t her first choice of food) they wouldn’t have received that money. Ben wouldn’t pay for emotional labor if there was no emotional labor to be found, and he wouldn’t pay them money to write an article about how Alice wanted Panda Express faster or how she felt that making over $100,000 a year was “tiny pay”.
Thanks. I don’t see any confirmation from him of actually offering to pay upfront, so barring that further evidence I would not read anything too definite from this.
If he says that he might pay them if he considers it to be sufficiently emotionally difficult for them, it still has the same incentive effects. If anything, an uncertain reward is more motivating and distorting than a certain one.
Especially since it seems likely that Alice tends to tell falsehoods when it will get her money. See here and here. Also, on priors, one of the most common reasons to lie is to get money.
Sorry I wasn’t clear. I mean that I haven’t seen him confirm publicly that he told them that he will or might pay them. The place you linked just talks about his draft plan of what he was thinking of doing (offering money). If he didn’t offer money to them, and they had no other indirect indication (until the process was over) that he was going to give them money, then there would be very little distortionary effect.
Oh, you’re right! I misread. I’ll update my comment to be more accurate.
Although I do think it’s decent odds that if he said that his plan was to discuss whistleblowing fees with them then, that he probably did. But it is much weaker evidence than I originally thought and conveyed.
I’ll concede that your comment which I criticised at the time is coming off better now. I still feel like it was over-claiming, but as I said, I’m now more sympathetic to it.
I agree with @Habryka that this comment is underspecified and likely written without proper review of the appendix linked. I suspect many readers are likely to conflate disputed with debunked, and this comment plays into that. This works so well, and it’s use is so widespread, that it has a name, FUD.
In the comments below, I have asked Spencer Greenberg to specify the most important claims he feels have been repudiated, and why he thinks so. I expect the answer will be genuinely elucidating to me.
ElliotJDavies—I had read earlier versions of the post and the appendix, which is why I felt somewhat confident in commenting on the quality of Ben Pace’s fact-checking (or lack thereof).
I think it is too early to make a call on whether Ben could have done a better job fact-checking things. The linked doc is 135 pages long, and I don’t think you or anyone else really has had time to read it and see how well it actually responds to the specific accusations.
I do think this is a pretty high-stakes situation, and people should be held accountable if they caused harm, but I think you are jumping to conclusions here at a speed that I don’t think can be well-justified.
I am myself still reading the 135 page document and trying to piece together which exact parts of the original post now seem inaccurate to me, and also whether Ben successfully communicated his epistemic state and seemed well-calibrated about the trustworthiness of different claims, which is an operation that I expect will take even very motivated people working on this-full time many hours. Over the coming days, I expect many people to read the linked documents, and piece together which claims have been responded to, which of the evidence provided is easily verifiable vs. hearsay, etc., and I think then we’ll have a much better guess whether there was some problem with fact-checking or accuracy of the original investigation.
Habryka—I read several earlier versions of the documentation. I’m familiar with the contents. I understand that others will want to take their time before reaching judgments. Fair enough.
Ah, that makes more sense. I agree that if you’ve read earlier versions of the documentation, then you are of course in a better position to judge things.
I do think this sentence could use some rephrasing in that case:
I hope it’s clear to most EAs that Ben Pace could have done a much better job fact-checking his allegations against Nonlinear, and in getting their side of the story.
Since you are talking about the epistemic state of “most EAs” here, who of course haven’t had advance access to the documentation here, and so aren’t really in a position to have made up their mind on this. I do think it’s fine for you to express a more overall judgement here, and am looking forward to comparing notes and takes when I am in a more similar epistemic state.
Whatever people think about this particular reply by Nonlinear, I hope it’s clear to most EAs that Ben Pace could have done a much better job fact-checking his allegations against Nonlinear, and in getting their side of the story.
In my comment on Ben Pace’s original post 3 months ago, I argued that EAs & Rationalists are not typically trained as investigative journalists, and we should be very careful when we try to do investigative journalism—an epistemically and ethically very complex and challenging profession, which typically requires years of training and experience—including many experiences of getting taken in by individuals and allegations that seemed credible at first, but that proved, on further investigation, to have been false, exaggerated, incoherent, and/or vengeful.
EAs pride ourselves on our skepticism and our epistemic standards when we’re identifying large-scope, neglected, tractable causes areas to support, and when we’re evaluating different policies and interventions to promote sentient well-being. But those EA skills overlap very little with the kinds of investigative journalism skills required to figure out who’s really telling the truth, in contexts involving disgruntled ex-employees versus their former managers and colleagues.
EA epistemics are well suited to the domains of science and policy. We’re often not as savvy when it comes to interpersonal relationships and human psychology—which is the relevant domain here.
In my opinion, Mr. Pace did a rather poor job of playing the investigative journalism role, insofar as most of the facts and claims and perspectives posted by Kat Woods here were not even included or addressed by Ben Pace.
I think in the future, EAs making serious allegations about particular individuals or organizations should be held to a pretty high standard of doing their due diligence, fact-checking their claims with all relevant parties, showing patience and maturity before publishing their investigations, and expecting that they will be held accountable for any serious errors and omissions that they make.
One area where Ben didn’t follow investigative journalism “best practices” (that I had missed early on, but saw mentioned in Kat’s post, and went back and checked) was that he financially compensated his sources ($5,000 each, or $10,000 total). This is frowned upon pretty heavily in investigative journalism (see e.g., https://en.wikipedia.org/wiki/Chequebook_journalism). I don’t have any reason to believe this meaningfully distorted the outcomes here (for instance, if the sources had no indication right until the end that Ben would compensate them financially, it is unlikely to have influenced their behavior) but it is a clear departure from an existing norm in the investigative journalism field. I appreciate that Ben disclosed this information; disclosure does address some but not all of the concerns around compensating sources.
I don’t rule out the possibility that the investigative journalism norm against paying sources is flawed, or it doesn’t apply in this case, or that a different set of norms should be applied.
Hmm, this seems like a pretty weak norm. In-particular the Wikipedia article you link says:
And I don’t have a sense that European investigative journalism is worse than U.S. investigative journalism.
Separately, whistleblower prices are quite common in the U.S. as well, for example: https://www.sec.gov/whistleblower
The linked Wikipedia article also has many quotes saying that really the central problem here is disclosure:
Habryka—Note that when the Securities & Exchange Commission (SEC) offers substantial whistleblower bounties, they do not simply take the whistleblowers’ word for it and start arresting people. They apply huge teams of auditors, lawyers, investigators, and federal agents to see if the whistleblower allegations are good enough to warrant legal action. If in doubt, they might convene a grand jury to see if the evidence is strong enough to take to trial. And they know from bitter experience that if they simply offered bounties to anybody who makes allegations, they would be deluged with false accusations.
Likewise with journalists. Yes, they offer payments in some cases for whistleblowers. But good journalists fact-check, with the expectation that many ‘whistleblowers’ will turn out to be bad actors with dubious agendas.
If we incentivize accusations, we’ll get a lot of false accusations. There has to be some good-faith effort to check if they are actually true.
Yep, I agree that it’s ideal for the prizes to be paid out conditional on them passing fact-checking or leading to a conviction. I think that was hard for various reasons in this case, but it seems clearly better for incentives.
I do think there was quite substantial good-faith effort of fact-checking involved. It might still be the case that it failed, I am still reading the giant 135-page document, but as many have pointed out, this investigation involved hundreds of hours of effort and interviews with over a dozen people, dozens of drafts, and many many fact-checks. So I do object to you implying there wasn’t any such process.
I think implying that the process that was present wasn’t enough and there should have been more, or that there were substantial issues with it, seems like a reasonable critique that I am still thinking about, but I think implying the absence of one seem bad.
My key point about investigative journalist expertise is that amateurs can invest a huge amount of time, money, and effort into investigations that are not actually very effective, fair, constructive, or epistemically sound.
As EAs know, charities vary hugely in their cost effectiveness. Investigative activities can also vary hugely in their time-effectiveness.
Yeah, I can totally imagine there are skills here that make someone substantially more effective at this (I think I have gotten vastly better at this skillset over the last 10 years, for example). As I said, I think criticizing the process seems pretty reasonable, I highly doubt that we went about this in the most optimal way.
Yep, I think we’re in accord on this.
I’m glad you’re taking time to read through the evidence and think on things.
For fact-checking, a basic thing would have been to speak to us and see our evidence. Like, we had interview transcripts and text messages and lot of relevant things. We reached out to Ben multiple times to talk to him and share this.
Ben updated a lot when he spoke to us. There was a lot of things they had told him that turned out to be false. He had reason to think we had more evidence like this, like what we’d already told him about and shared.
He also had all of the information needed to know that we weren’t the retaliatory villains Alice and Chloe painted us out to be, who would attack them for sharing their side. They had been telling bad thing about us for 1.5 years and none of their fears came true.
The only thing we ever did was share our side, which they tried to portray as unethical.
It’s not that Ben didn’t have a process. It’s that his process would predictably lead to incorrect conclusions. Spending less than 1% of the time talking to and trying to understand the other side and spending 99% of the time on one side is soldier mindset, not scout.
I continue to think that our concerns about retaliation were well-placed. Ben did talk to you multiple times, which I think clears the basic bar for due-diligence.
Sadly due to the track record of retaliation that you do seem to have displayed, I continue to endorse not engaging with you further during the investigation, though maybe I will change my mind on that after reading more of the evidence document.
I really wish things were different and we could have collaboratively investigated the accusations, but man, yeah, the libel threat was really bad, as were a bunch of other things that we heard about you and you said to us directly, and also of course we were concerned about retaliation to our sources and didn’t see a way to avoid exposing them to more risk from you without having things public.
There isn’t a track record of retaliation. We didn’t retaliate against your sources. We know who almost all of them are and what they said and nothing happened to them.
Alice’s messages simply show me saying that if she continued sharing her side, I would share mine. Sharing your side is not unethical.
And the examples that people gave of retaliation for Emerson were of him being sued and people sharing their side online, and him replying saying he’s countersue and he’d share his side (which he hadn’t done yet). This isn’t unethical, but a very reasonable thing to do.
For the libel, Ben knowingly said multiple things that were false and damaging, and he said dozens of things that he could have easily known were false if he’d just waited a week out of 6 months.
But we never wanted to sue Ben. We just wanted Ben to give us time to look at the evidence we were more than willing to share with him. I really recommend reading this section, because I think it gets across very well what was happening.
Here’s a quick excerpt:
And the accusation of threatening to hire stalkers is just a really weird accusation. That should be an indicator that somebody is not mentally alright.
I’m really sad too that we couldn’t just talk too. I hope we still might be able to, once you’ve read the document and see that the retaliation reputation was unwarranted. I would really love to talk. I think trying to do conflict resolution in a high stakes, hostile, and public venue is less likely to work than if we can talk face-to-face and have a higher bandwidth conversation.
Honestly, I wish we’d already invented mind reading technology, because I’d just let you read my mind, unfiltered. I know that if you could, you’d see that I really have no negative intentions and I’m really just trying to figure out how to make everybody happy and reduce suffering. This situation is complicated and I certainly can sometimes unintentially cause harm, and I hate that, and I’m always working on trying to prevent that. But I really do just want everybody to be happy, including you. Anyways, for now we don’t have mind-reading technology that’s accurate or cheap enough, so we’ll have to make do with me trying to convey through text that I really am not retaliatory. If you hurt me, I will try to understand you, try to help you understand me, then try to collaboratively problem-solve.
My original comment left a pretty wide window of possibilities open, and your reply falls within that window, so I don’t quite think we disagree a lot. However, in the spirit of nitpicking, I’ll make a couple of points:
Prominence of disclosure matters. The fact that Ben included the information in his post shows that he didn’t intend to hide it; nonetheless, my sense is that he didn’t highlight it as a disclosure / disclaimer / caveat for readers to keep in mind when interpreting the post. He did include other disclaimers around his process and motivation at the start of the post, that I found helpful, and his non-inclusion of payment along with those disclosures gives me the sense that he didn’t consider the distortionary effect of payment as a biasing factor worth highlighting to readers. My guess is that it would be pretty likely for readers to miss it (as I did). I’m genuinely uncertain whether the lack of discussion around this was driven by people not noticing it, or noticing it and not thinking it mattered.
I’m familiar with the broad outlines of the whistleblower law (from this podcast episode). I think there’s a distinction, though, between awarding money after a determination / judgment of harm, versus awarding money as a journalist or investigator who’s trying to report on the situation. I don’t know exactly how Ben perceived his role, and perhaps the point is that he didn’t perceive his role as being strictly one or the other, but a mix.
Yeah, this makes sense. FWIW, my current memory of the situation was that Ben hadn’t made any promises about paying for information until quite late in the process, and the primary purpose of the payment was to enable the publishing of information that was already circling around privately (i.e. in private docs that Alice and Chloe had shared with some others).
Of course, it’s hard to get rid of the incentive, but I think given that it was paying for publishing something that was already largely written up, I do think the immediate incentives here are weaker (though of course in the long run, and also via various more TDT-ish considerations, there is still an effect here).
I also am not super confident in the exact historical details here. Slack records suggest the rewards hadn’t been finalized the week before the post.
Thanks. Do you remember when Ben started discussing the possibility of pay?
Alas, I do not know. I have some internal Slack records suggesting it as a thing to consider in April, but I don’t know when Ben brought it up to Alice or Chloe. I am confident nothing was confirmed until quite close to the post being published, but I don’t know when the idea was first floated (with the only bound I have is that it probably wasn’t before April).
“High-quality information” is key. Ben did not fact check basic things and we’ve provided evidence that a huge amount of the “witnesses testimony” was false or misleading.
If Ben had waited one week out of the 6 months he spent working on this, he would have known this and not trusted the sources. They said dozens of provably false things and we just wanted some time to share it all with him, since they’d accused us of so many things.
We didn’t want ages—we just wanted a week. And Ben had been working on it for 6 months, so it didn’t seem like that much to ask.
Paying people $10,000 to say untrue or misleading information seems bad. People should not be paid until their facts are checked, and if it’s shown that their facts were false, they should not be paid.
Agree that in as much as people were paid directly for propagating inaccurate information, then that seems sad and clearly sets the wrong incentives. I am not currently convinced of that after the initial reading of your evidence document, but I am still reading, and there really is a lot of stuff to process.
In this thread I am trying to have a locally valid discussion on the actual presence of norms against paying for information among investigative journalists. I would prefer if we can keep the discussion here more local since it seems like an interesting and somewhat important question, and I think it would be an important update for me if there were was a consensus among investigative journalists and similar professions that whistleblower prizes and paying for information is a bad idea.
vipulnaik—good point.
This is also why I would be very wary of the EA Community Safety team offering ‘whistleblower support’ (which could boil down to ‘bounties for false accusations’).
Ben says that he was discussing offering it to them months before publishing. [EDIT: he didn’t say he did discuss this with them. He just said he planned to.]
I think it does incentivize them to distort what they say. They were incentivized to make everything sound maximally sad-sounding. Ben said if they did the emotional labor of sharing their sad stories, he’d give them $10,000.
They knew that if their stories hadn’t been very sad (e.g. Alice said she did get food but it just wasn’t her first choice of food) they wouldn’t have received that money. Ben wouldn’t pay for emotional labor if there was no emotional labor to be found, and he wouldn’t pay them money to write an article about how Alice wanted Panda Express faster or how she felt that making over $100,000 a year was “tiny pay”.
Thanks. I don’t see any confirmation from him of actually offering to pay upfront, so barring that further evidence I would not read anything too definite from this.
If he says that he might pay them if he considers it to be sufficiently emotionally difficult for them, it still has the same incentive effects. If anything, an uncertain reward is more motivating and distorting than a certain one.
Especially since it seems likely that Alice tends to tell falsehoods when it will get her money. See here and here. Also, on priors, one of the most common reasons to lie is to get money.
Sorry I wasn’t clear. I mean that I haven’t seen him confirm publicly that he told them that he will or might pay them. The place you linked just talks about his draft plan of what he was thinking of doing (offering money). If he didn’t offer money to them, and they had no other indirect indication (until the process was over) that he was going to give them money, then there would be very little distortionary effect.
Oh, you’re right! I misread. I’ll update my comment to be more accurate.
Although I do think it’s decent odds that if he said that his plan was to discuss whistleblowing fees with them then, that he probably did. But it is much weaker evidence than I originally thought and conveyed.
Yes, when I saw that, I had to wonder whether the payment was offered afterward (as a gift) or in advance (possibly in exchange for information).
(It was offered afterwards)
I’ll concede that your comment which I criticised at the time is coming off better now. I still feel like it was over-claiming, but as I said, I’m now more sympathetic to it.
I agree with @Habryka that this comment is underspecified and likely written without proper review of the appendix linked. I suspect many readers are likely to conflate disputed with debunked, and this comment plays into that. This works so well, and it’s use is so widespread, that it has a name, FUD.
In the comments below, I have asked Spencer Greenberg to specify the most important claims he feels have been repudiated, and why he thinks so. I expect the answer will be genuinely elucidating to me.
ElliotJDavies—I had read earlier versions of the post and the appendix, which is why I felt somewhat confident in commenting on the quality of Ben Pace’s fact-checking (or lack thereof).
I think it is too early to make a call on whether Ben could have done a better job fact-checking things. The linked doc is 135 pages long, and I don’t think you or anyone else really has had time to read it and see how well it actually responds to the specific accusations.
I do think this is a pretty high-stakes situation, and people should be held accountable if they caused harm, but I think you are jumping to conclusions here at a speed that I don’t think can be well-justified.
I am myself still reading the 135 page document and trying to piece together which exact parts of the original post now seem inaccurate to me, and also whether Ben successfully communicated his epistemic state and seemed well-calibrated about the trustworthiness of different claims, which is an operation that I expect will take even very motivated people working on this-full time many hours. Over the coming days, I expect many people to read the linked documents, and piece together which claims have been responded to, which of the evidence provided is easily verifiable vs. hearsay, etc., and I think then we’ll have a much better guess whether there was some problem with fact-checking or accuracy of the original investigation.
Habryka—I read several earlier versions of the documentation. I’m familiar with the contents. I understand that others will want to take their time before reaching judgments. Fair enough.
Ah, that makes more sense. I agree that if you’ve read earlier versions of the documentation, then you are of course in a better position to judge things.
I do think this sentence could use some rephrasing in that case:
Since you are talking about the epistemic state of “most EAs” here, who of course haven’t had advance access to the documentation here, and so aren’t really in a position to have made up their mind on this. I do think it’s fine for you to express a more overall judgement here, and am looking forward to comparing notes and takes when I am in a more similar epistemic state.
Yes, I should have said ‘I hope it will be clear to most EAs that Ben Pace could have done a much better job....’