Some post-EAG thoughts on journalists
For context, CEA accepted at EAG Bay Area 2023 a journalist who has at times written critically of EA and individual EAs, and who is very much not a community member. I am deliberately not naming the journalist, because they haven't done anything wrong and I'm still trying to work out my own thoughts.
On one hand, "journalists who write nice things get to go to the events, journalists who write mean things get excluded" is at best ethically problematic. It's very very very normal: political campaigns do it, industry events do it, individuals do it. "Access journalism" is the norm more than it is the exception. But that doesn't mean that we should. One solution is to be very very careful about maintaining the differentiation between "community member" and "critical or not". Dylan Matthews is straightforwardly an EA and has reported critically on a past EAG: if he was excluded for this I would be deeply concerned.
On the other hand, I think that, when hosting an EA event, an EA organization has certain obligations to the people at that event. One of them is protecting their safety and privacy. EAs who are journalists can, I think, generally be relied upon to be fair and to respect the privacy of individuals. That is not a trust I extend to journalists who are not community members: the linked example is particularly egregious, but tabloid reporting happens.
EAG is a gathering of community members. People go to advance their goals: see friends, network, be networked at, give advice, get advice, learn interesting things, and more. In a healthy movement, I think that EAGs should be a professional obligation, good for the individual, or fun for the individual. It doesn't have to be all of them, but it shouldn't harm them on any axis.
Someone might be out about being bi at an after-party with friends, but not want to see that detail being confirmed by a fact-checker for a national paper. This doesn't seem particularly unusual. They would be right to trust community members, but might not realize that there could be journalists at the after-party. Non-community journalists will not necessarily share norms about privacy or have particularly strong incentives to follow any norms that do exist.
On the gripping hand, it feels more than a little hypocritical to complain about the low quality of criticism of EA and also complain when a journalist wants to attend an EA event to get to know the movement better.
One thing I'm confident of is that I wish that this had been more clearly disclosed. "This year we are excited to welcome X, who will be providing a critical view on EA" is good enough to at least warn people that someone whose bio says that they are interested inÂ
how the wealthiest people in society spend their money or live their lives
(emphasis mine)
is attending.
I'm still trying to sort out the rest of my views here. Happy to take feedback. It's very possible that I'm missing some information about this.
P.S.
I have been told by someone at CEA that all attending journalists have agreed that everything at EAG is off the record by default. I don't consider this to be an adequate mitigating factor for accepting non-community journalists and not mentioning this to attendees or speakers.
P.P.S.
And no, I'm not using a pseudonym for this. I think that that is a bad and damaging trend on the Forum, and I don't, actually, believe that anyone at CEA will retaliate against me for posting this.
Bad Things Are Bad: A Short List of Common Views Among EAs
Hope that clears everything up. I expect with 90% confidence that over 90% of EAs would agree with every item on this list.
Inside, I don't know enough to say with confidence. Could be caste discrimination, could be ongoing oppression of non-Han, could be something I'm not thinking of. I'm not making a claim about the globe as a whole because I haven't run the numbers, and different EAs will have different values and approaches to how to weight history, cultures, etc. I just refuse to fall into the standard America/Euro-centric framework.
I might change 2 to:
No, we should not murder AI researchers. Working together requires ability to trust one another. Hurting AI researchers will just damage this trust and likely not slow down the work. Likewise, if you are the sort of person who thinks you should do this, you are likely uniquely unsuited to coming up with drastic solutions. Have you tried a research agenda? If not, why did you start at murder. We are all more likely to survive if we can credibly commit not to defect. Please get help.
I’m concerned that less than 90% of the AI safety community would agree. I have heard some disturbing anecdotes.