Comment by dwohnitmok
2 months ago
> I suspect you think more effort went into my comment than actually did. I spent less than 60 seconds on: clicking two or three buttons, typing out the names I saw from the other window, then scrolling down and seeing the 501(c)3.
This is a bit frustrating of a response to get. No, I don't believe you spent a lot of time on this. I wasn't imaging you spending hours or even minutes tracking these guys down. But I also don't think it's relevant.
I don't think you'd find it relevant if the Sage researchers said "I didn't spend any effort on this. I only did this because I wanted to make the point that AIs have enough capability to navigate the web and email people. I could have made the point abstractly, but abstract arguments are harder to follow. There was no other intention than what I put in the prompt." It's hence frustrating to see you use essentially the same thing as a shield.
Look, I'm not here to crucify you for this. I don't think you're a bad person. And this isn't even that bad in the grand scheme of things. It's just that naming and shaming specific people feels like an overreaction to thoughtless, machine-generated thank you emails.
I went for a walk to think about your position. I do not think you are wrong. If you refused to name a person in a situation like this, I would never try to convince you otherwise. That is why it is hard for me to make a case to you here, because I do not hold the opposing position. But I also find your argument that I should have not done so unconvincing. Both seem like reasonable choices to me.
I have two tests for this. First: what harm does my comment here cause? Perhaps some mild embarrassment? It could not realistically do more.
Second: if it were me, would I mind it being done to me? No. It is not a big deal. It is public feedback about an insulting computer program, no one was injured, no safety-critical system compromised. I have been called out for mistakes before, in classes, on mailing lists, on forums, I learn and try to do better. The only times I have resented it are when I think the complaint is wrong. (And with age, I would say the only correct thing to do then is, after taking the time to consider it carefully, clearly respond to feedback you disagree with.)
The only thing I can draw from thinking through this is, because the authors of the program probably didn’t see my comment, it was not effective, and so I would have been better emailing them. But that is a statement about effectiveness not rightness. I would be more than happy doing it in a group in person at a party or a classroom. Mistakes do not have to be handled privately.
I am sorry we disagree about this. If you think I am missing anything I am open to thinking about it more.
> I am sorry we disagree about this. If you think I am missing anything I am open to thinking about it more.
I am sorry I'm responding to this so late. I very much appreciate the dialogue you're extending here! I don't think I'll have the time to give you the response you deserve, but I'll try to sketch out some of the ideas.
This is all a matter of degree. Calling individuals out on mailing lists, in internal company comms, or in class still feels different than going and listing all an org's members on a website (even more so than e.g. just listing the CEO).
There's a couple of factors here at play, but mainly it's the combination of:
1. The overall AI trend is a large, impactful thing, but this was a small thing 2. Just listing the names without any explanation other than "they're responsible"
This just pattern matches to types of online behavior I find quite damaging for discourse too closely for my liking.