When the board of OpenAI moved to fire Sam Altman, much of Twitter leaped to his defense. Their narrative was that the board was composed of busybody know-nothings committed to abstract ideals, poised to sink a company with the potential to deliver real growth. The terse statement of the board that Altman’s behavior had raised questions of honesty attracted particular ire. The board members were subject to vitriolic criticism and at the risk of sounding precious, some of it was in my view quite sexist. For example, it was alleged that Helen Toner was not qualified for the board, because she had relatively little academic industry experience, and her highest qualification was a master's in security studies. What this ignored, among other things, was her impressive record of journal articles and other publications.
"An ursine Leviathan, its halo studded with a range of arcane symbols, including one snowflake and several sea urchin fossils, looming over a naval battle fought by yachts, dhows, junks, and moose-riders among the surprisingly well-preserved ruins of a flooded London, while three-winged eldritch horrors circle overhead. Dichromatic woodcut."
Haha! It was actually " There's a classic image on the cover of Leviathan by Thomas Hobbes, generate me that image except the king is a bear, 5:3 aspect ratio"
It's almost impossible to know enough through testimony to make an ethical judgement about complex interpersonal matters with any reliability. Not impossible to know enough, but insanely difficult in a way that isn't normally recognised by internet culture (and culture generally). By comparison, policy is much easier.
You wouldn’t make an infallible judgement of course. You’d make a provisional judgement. We have to do this all the time. If you were deciding whether to take a class with a professor and a bunch of people told you this professor was an unfair grader, you would believe them, for example.
We always have to use testimony about ‘complex interpersonal matters’ as well as about the character of other people, what they did, how they might have been motivated, what to expect from them, and so on.
How could we even function if we did not do this sometimes? We cannot witness every event. How would history even be possible. It’s all testimony.
To be honest, if a bunch of my friends said someone was an unfair grader, I'd probably act when I'm around them on the assumption their accusation is true, so they felt supported and validitated, but privately reserve judgement. This might change if there were undeniable signs, like the marker openly writing "I have marked out down because I hate this theory", but for the most part, I'd be uncertain of any claim, internally. Now I might avoid the class, but that's very far from the standard of knowledge, which is what I'm mostly concerned with here. It's also below the standard of something understood confidently enough to talk about publicly, especially where mistakes can be costly.
I truly think people vastly overestimate how much they know about the social sphere, even the parts of it they have contact with, and should talk about interpersonal relationships far less confidently than we do (and I include myself in that- I am often way too confident). Exactly how to quantify or state this more formally, I am not sure.
But wouldn’t it be the standard you would use to judge any interpersonal situation where your evidence involves relying on other people?
You would of course know that there are many factors that could make their claim untrue or unfair.
It would be a provisional belief you would act on because you have some evidence but not all the evidence.
We have to do that. We are continually making these judgements. It’s understood there is only a chance they are correct.
That’s all I mean. If a number of people give plausible overlapping testimony about an interpersonal situation or a person and there is enough reason to think they are reliable then you may also make a judgement.
Of course you would not bet the farm on it. You may only feel confident at best, not completely certain
It’s simply something we do because we have to do it. Our safety and welfare and relationships and so on may require us to do it.
"An ursine Leviathan, its halo studded with a range of arcane symbols, including one snowflake and several sea urchin fossils, looming over a naval battle fought by yachts, dhows, junks, and moose-riders among the surprisingly well-preserved ruins of a flooded London, while three-winged eldritch horrors circle overhead. Dichromatic woodcut."
Haha! It was actually " There's a classic image on the cover of Leviathan by Thomas Hobbes, generate me that image except the king is a bear, 5:3 aspect ratio"
But we know all about Kate and Meghan...
I am perplexed by the 'you don't know them' claim.
What is it supposed to mean.
We generally make judgements about many people through testimony. Is all testimony unreliable?
It's almost impossible to know enough through testimony to make an ethical judgement about complex interpersonal matters with any reliability. Not impossible to know enough, but insanely difficult in a way that isn't normally recognised by internet culture (and culture generally). By comparison, policy is much easier.
You wouldn’t make an infallible judgement of course. You’d make a provisional judgement. We have to do this all the time. If you were deciding whether to take a class with a professor and a bunch of people told you this professor was an unfair grader, you would believe them, for example.
We always have to use testimony about ‘complex interpersonal matters’ as well as about the character of other people, what they did, how they might have been motivated, what to expect from them, and so on.
How could we even function if we did not do this sometimes? We cannot witness every event. How would history even be possible. It’s all testimony.
To be honest, if a bunch of my friends said someone was an unfair grader, I'd probably act when I'm around them on the assumption their accusation is true, so they felt supported and validitated, but privately reserve judgement. This might change if there were undeniable signs, like the marker openly writing "I have marked out down because I hate this theory", but for the most part, I'd be uncertain of any claim, internally. Now I might avoid the class, but that's very far from the standard of knowledge, which is what I'm mostly concerned with here. It's also below the standard of something understood confidently enough to talk about publicly, especially where mistakes can be costly.
I truly think people vastly overestimate how much they know about the social sphere, even the parts of it they have contact with, and should talk about interpersonal relationships far less confidently than we do (and I include myself in that- I am often way too confident). Exactly how to quantify or state this more formally, I am not sure.
But wouldn’t it be the standard you would use to judge any interpersonal situation where your evidence involves relying on other people?
You would of course know that there are many factors that could make their claim untrue or unfair.
It would be a provisional belief you would act on because you have some evidence but not all the evidence.
We have to do that. We are continually making these judgements. It’s understood there is only a chance they are correct.
That’s all I mean. If a number of people give plausible overlapping testimony about an interpersonal situation or a person and there is enough reason to think they are reliable then you may also make a judgement.
Of course you would not bet the farm on it. You may only feel confident at best, not completely certain
It’s simply something we do because we have to do it. Our safety and welfare and relationships and so on may require us to do it.
That’s all I mean.
AI will never be more “intelligent” than the people with their fingers on the scales.