I think this is one of those places where a relative outsider may have a comparative advantage in some cases. Speaking for myself, I think I would be less likely to blurt if my employer had anything to do with EA, if I was hoping to get grants, if my sense of social identity was strongly tied to the EA community, etc. Of course, it means I know less and am more likely to say something stupid, too . . .
As a thought experiment, and recognizing the huge risk of hindsight bias, consider if a well-educated and well-informed outsider had been asked to come up a year ago with a list of events that would have been moderately serious to catastrophic for EA and estimated probabilities. I think there is a very high chance they would have come up with “loss of a megadonor,” and a high chance they would have come up with “and I note that SBF’s assets are in a high volatility field with frequent bankruptcies, in a very young company.”I think there is a very high chance that they would have come up with “major public scandal involving a key figure,” and a high chance they would have come up with a list of key figures that included SBF. I suspect that most outsiders would have assigned a higher probability to these things than most insiders, although this is impossible to test at this point.
I don’t think it likely “SBF is discovered to be running a massive fraud” would explicitly be on most hypothetical outsiders’ list, but I think it is more likely to have appeared than if an insider had prepared the list. That is due to (1) self-censoring, and (2) less deference to / weighing of what key figures in the community thought of SBF. At least when you’re talking about assessing serious to catastrophic risks, I think you want to get both the “insider” and “outsider” views.
I think this is one of those places where a relative outsider may have a comparative advantage in some cases. Speaking for myself, I think I would be less likely to blurt if my employer had anything to do with EA, if I was hoping to get grants, if my sense of social identity was strongly tied to the EA community, etc. Of course, it means I know less and am more likely to say something stupid, too . . .
As a thought experiment, and recognizing the huge risk of hindsight bias, consider if a well-educated and well-informed outsider had been asked to come up a year ago with a list of events that would have been moderately serious to catastrophic for EA and estimated probabilities. I think there is a very high chance they would have come up with “loss of a megadonor,” and a high chance they would have come up with “and I note that SBF’s assets are in a high volatility field with frequent bankruptcies, in a very young company.”I think there is a very high chance that they would have come up with “major public scandal involving a key figure,” and a high chance they would have come up with a list of key figures that included SBF. I suspect that most outsiders would have assigned a higher probability to these things than most insiders, although this is impossible to test at this point.
I don’t think it likely “SBF is discovered to be running a massive fraud” would explicitly be on most hypothetical outsiders’ list, but I think it is more likely to have appeared than if an insider had prepared the list. That is due to (1) self-censoring, and (2) less deference to / weighing of what key figures in the community thought of SBF. At least when you’re talking about assessing serious to catastrophic risks, I think you want to get both the “insider” and “outsider” views.