Latest / Ethical Machines / We Are All Responsible for AI, Part 2

Transcript

We Are All Responsible for AI, Part 2 · Ethical Machines

0:01
[SPEAKER_01]: A couple of weeks ago, I released an episode with Brian Wong, who's a professor of philosophy at the University of Hong Kong, and we got into talking about the ways in which AI causes harm to lots of people.
0:12
[SPEAKER_01]: And more specifically, we talked about these two related concepts that he has, the accountability gap, and the accountability deficit.
0:19
[SPEAKER_01]: the accountability gap, which I know you recall from the last episode, has to do with the fact that lots of people get hurt by AI, but there's no discrete or particular individuals who are responsible for it.
0:30
[SPEAKER_01]: It's not, oh, we did that to them.
0:32
[SPEAKER_01]: Rather, it's if you like the AI industrial complex that led to that kind of harm to people.
0:37
[SPEAKER_01]: So there's a gap because it's not clear who was supposed to hold accountable.
0:43
[SPEAKER_01]: The accountability deficit is just what happens when no one, if you like, pays up.
0:48
[SPEAKER_01]: The people are at harm, they're owed it.
0:50
[SPEAKER_01]: Looks like they're owed some kind of reparations, but it's not clear who owes them what.
0:54
[SPEAKER_01]: So that's the accountability gap and the accountability deficit.
0:58
[SPEAKER_01]: And what that leads to brine to is the following claim.
1:01
[SPEAKER_01]: That we're all responsible in somebody.
1:02
[SPEAKER_01]: We're all responsible for the harms that have visited upon these people.
1:07
[SPEAKER_01]: And it's a bit of an odd position, though, because he thinks that
1:11
[SPEAKER_01]: even though people are responsible, you and I are responsible for that harm were not blameworthy in any particular way.
1:19
[SPEAKER_01]: So let me say something about that to make sure that you are trying to why this is so philosophically substantive.
1:24
[SPEAKER_01]: Usually we think that if someone is
1:28
[SPEAKER_01]: then if they don't do that thing for which they're morally responsible, then they deserve moral blame.
1:33
[SPEAKER_01]: So for instance, I'm morally responsible for taking care of my children, making sure that they're well fed, they have a house, blah, blah, blah.
1:39
[SPEAKER_01]: And if I don't take care of them in those ways, then I'm blame worthy.
1:43
[SPEAKER_01]: I should be blamed for not taking good care of them.
1:46
[SPEAKER_01]: If I go off and I steal something from someone, not to feed my children or anything like that, but just for fun or because I want the thing, then I've breached a moral obligation to not steal
1:57
[SPEAKER_01]: So this is a standard way that philosophers and most people think about responsibility and blame.
2:03
[SPEAKER_01]: Each individual or individual has a responsibility to do or turn out to do something and failing to do the thing or doing nothing they weren't supposed to do means that they deserve to be blamed.
2:14
[SPEAKER_01]: But what's interesting about Bryan's position is that he thinks that there's a kind of way in which we can have a moral responsibility and failing to love up to that moral responsibility doesn't entail that we're being blamed worthy.
2:29
[SPEAKER_01]: So we are responsible.
2:31
[SPEAKER_01]: We the collective engage in the AI industrial complex.
2:34
[SPEAKER_01]: Our responsible for those harms that are visited upon various people.
2:38
[SPEAKER_01]: But it's not the case that we're blame worthy in any way if we fail to discharge whatever our responsibilities are.
2:45
[SPEAKER_01]: So this is what this part two conversation is about now.
2:48
[SPEAKER_01]: What is that really amount to?
2:50
[SPEAKER_01]: You can't just say, oh, it would be nice of us if we helped out those people that were harmed.
2:56
[SPEAKER_01]: But because it's not enough to say it on Brian's account anyway, it's not enough to say,
3:00
[SPEAKER_01]: It would be nice of us to do it.
3:02
[SPEAKER_01]: It's not if you like over and above the call of duty.
3:04
[SPEAKER_01]: It is the call of duty.
3:05
[SPEAKER_01]: We are responsible for doing things that would repay those people who have been harmed and or do things that would stop future harms from being visited upon people.
3:17
[SPEAKER_01]: There's two things that I want to focus on, or two things that I want to highlight that you're going to hear a lot in the conversation.
3:21
[SPEAKER_01]: One is a question about what makes it the case that we're responsible for those harms.
3:27
[SPEAKER_01]: And there, Brian's position is something like this.
3:30
[SPEAKER_01]: When you and I engage in the AI industrial complex, as I've just been calling it, we use AI, we go on with the CHPT or onto cloud.
3:38
[SPEAKER_01]: We use AI in various ways.
3:40
[SPEAKER_01]: We talk about a blah, blah, blah.
3:41
[SPEAKER_01]: We are.
3:43
[SPEAKER_01]: part of the background conditions that reproduce the harms that are getting done, we're part of the background conditions that reproduce the harms that are being done.
3:52
[SPEAKER_01]: And so that's what makes it the case.
3:55
[SPEAKER_01]: That's what grounds are what he calls forward looking responsibilities, forward looking responsibilities are these special responsibilities that we don't get in blame if we don't look up to them.
4:05
[SPEAKER_01]: So you guys are thinking I want to highlight it because it might help in some way as you listen to Brian articulated the position is this point that he makes that you can be criticized without being blamed you can be criticized without being blamed.
4:16
[SPEAKER_01]: So what does that look like?
4:17
[SPEAKER_01]: I'm not sure this is the best example, but this is what I use them in my conversation with Brian.
4:22
[SPEAKER_01]: you can criticize a kid for being a spoiled brat, or maybe there are five years old they're spoiled brat, they're deserving of some kind of criticism or you can critique them.
4:31
[SPEAKER_01]: But you don't blame them, it's not their fault that they're spoiled brat, they're parents or the ones that spoiled them.
4:35
[SPEAKER_01]: So you can criticize the kid or critique the kid as being a spoiled brat, perhaps not to his face, but that doesn't mean that you're blaming them.
4:43
[SPEAKER_01]: So I think that Brian wants to make use of this distinction where we can be critiqued or criticized for not discharging these special kinds of responsibilities, even though we can be blamed for it.
4:56
[SPEAKER_01]: Okay, there's lots of this conversation as you'll see what I want to highlight though and what I like so much about these two episodes, this is something I said last time as well, is that.
5:06
[SPEAKER_01]: were all familiar with the sort of big harms that are visited upon moths people by virtue of people using AI that are by rental harms the ways in which labor is exploited to build AI in the first place and we start getting into okay so these bad things are happening so who's responsible.
5:24
[SPEAKER_01]: who's accountable and what we're forced into intellectually speaking is thinking about what we mean when we say who's responsible?
5:32
[SPEAKER_01]: Is that known as responsible?
5:34
[SPEAKER_01]: Is that as Brian's Brian maintains where all responsible?
5:38
[SPEAKER_01]: But what is that really amount to?
5:39
[SPEAKER_01]: And so we go from these concrete harms visited upon people to sort of intellectual ethical inquiry
5:53
[SPEAKER_01]: The purpose and value of philosophical inquiry, because what we're doing in this conversation is really hardcore, deep, philosophical inquiry that kind of stuff that Professor Langeon, that professional philosophers engage in.
6:04
[SPEAKER_01]: So even if you don't follow every in and out of this argumentation that you're about to listen to, what I hope you come to appreciate is the way in which this kind of inquiry, this discussion, is called for given a concern for these very concrete harms.
6:21
[SPEAKER_01]: Okay, so big conversation, let's go on on.
6:23
[SPEAKER_01]: If you like this kind of stuff, then please consider supporting me at readblapin.substack.com.
6:28
[SPEAKER_01]: I think I charged now $5 a month or $60 a year.
6:31
[SPEAKER_01]: I don't even charge.
6:32
[SPEAKER_01]: I just say if you want to donate and you want to support the podcast that I appreciate it.
6:35
[SPEAKER_01]: To those of you who did subscribe, since the last episode, thank you so much.
6:39
[SPEAKER_01]: I appreciate it.
6:40
[SPEAKER_01]: If you want to email me with suggestions about who to have on the podcast, you've got questions, trenches, criticism, EM at readblapin.com.
6:46
[SPEAKER_01]: All right, let's go talk to Brian.
7:00
[UNKNOWN]: Thank you.
7:05
[SPEAKER_01]: All right, so Brian, last time we talked towards the end of our conversation, we started talking about these two concepts that you highlighted.
7:11
[SPEAKER_01]: One is the accountability gap, the other is the accountability deficit.
7:16
[SPEAKER_01]: So the accountability gap is bad things happen to some people.
7:19
[SPEAKER_01]: Let's say it's the result of AI or whatever.
7:21
[SPEAKER_01]: It doesn't really matter, but bad things happen.
7:23
[SPEAKER_01]: And there's no individual or set of individuals, the screen individuals who we can say, they're the ones we should hold accountable.
7:30
[SPEAKER_01]: Because there's too much of a crazy causal mix, I suppose, to say that's the person.
7:34
[SPEAKER_01]: And so that's the accountability gap.
7:37
[SPEAKER_01]: And then the deficit is bad things have happened to these people.
7:40
[SPEAKER_01]: They're owed some kind of compensation or reparations or something on those lines, just the justice demands that they get some kind of compensation.
7:48
[SPEAKER_01]: And yet they've received no such compensation and so that's the deficit.
7:52
[SPEAKER_01]: So,
7:54
[SPEAKER_01]: We then talked about, okay, so there's two claims here, you could think if we can't identify discrete individuals or certain individuals that we can hold accountable, then I guess no one's responsible, no one's accountable, and we're just to sort of, you didn't put it this way, but I'll put it this way.
8:10
[SPEAKER_01]: It's just a shitty situation, which bad things happen and no one's responsible.
8:13
[SPEAKER_01]: That's one option.
8:14
[SPEAKER_01]: But you reject that option.
8:15
[SPEAKER_01]: You go in for a different claim, which is if we can't identify discrete individuals for whom we should hold accountable for the bad things that happened, then everyone is responsible.
8:27
[SPEAKER_01]: And that's really interesting claim, and so I take it that the purpose of this conversation is to explore the underpinnings, the justification for what looks like a rather odd claim, at least I think on the Facebook, you will say, no, why should what did I have to do with it?
8:43
[SPEAKER_01]: So I think I think that's where we are now.
8:44
[SPEAKER_01]: So let's get into it.
8:46
[SPEAKER_01]: Why should we think that everyone is responsible when bad things happen for which there's no discrete individuals to hold accountable?
8:52
[SPEAKER_01]: Thank you very much.
8:53
[SPEAKER_01]: And let's, sorry, I'll say one last thing.
8:55
[SPEAKER_01]: Let's think that's super abstract.
8:57
[SPEAKER_01]: We're philosopher, so we speak in the other.
8:58
[SPEAKER_01]: But to XIX, we're more abstract, if there are some AI system out there, that's passed through the hands of dozens, maybe hundreds of data scientists and executives who have green-lit things at each stages and different people have used different tools and there's data collectors and there's people who have done the fine tuning and there's so many, and then you wind up with, let's say,
9:20
[SPEAKER_01]: widespread discrimination against, let's say women or people of color or something like that, or there's massive privacy violations.
9:29
[SPEAKER_01]: And we want to hold the screen individuals accountable.
9:32
[SPEAKER_01]: How do we do that?
9:32
[SPEAKER_01]: And your point is something like everyone is.
9:35
[SPEAKER_01]: So let's take it from there.
9:36
[SPEAKER_00]: Thanks Reed.
9:37
[SPEAKER_00]: I want to just start off by making a pretty obvious observation that perhaps runs slightly contrary to what I'm suggesting and advancing at large, which I think you've quite accurately captured.
9:46
[SPEAKER_00]: I don't think anyone at all is to be held responsible for light nicks, alternatively for rainstorms or tropical cyclones, probably so.
9:55
[SPEAKER_00]: Yeah, they have climate change and also run away green house gas and green house effects.
9:59
[SPEAKER_00]: You might argue that tropical side things are to do with human action.
10:02
[SPEAKER_00]: But by and large, natural disasters are not what would call some of that grounds retrospective liability, whether it be in a sort of blame-centred set for which someone's been where they, on a non-blamer touch set, so I would still wouldn't say, hey, you're read responsible for the rainstorm.
10:17
[SPEAKER_00]: So I want to just sort of consider conceding and noting that I'm not suggesting that there's something surely abstract purely natural and ultimately uncaused by human action that we're still going to say everyone's liable, what is very least to the extent of liability, it's no more relevant and no more than just merely a general duty to mitigate the harms of tropical cyclones by both anti flooding barriers and architectures.
10:39
[SPEAKER_00]: That's the first piece of my response to question.
10:45
[SPEAKER_00]: by saying that agent X has a degree of liability.
10:49
[SPEAKER_00]: That all that means is that it should be open to taking up a shortening portion of the costs required, and that could be a monetary, it could be non-monentary, in redressing something, or alternatively, in undertaking something, okay, so take a step back and say in undertaking some omega or some delta.
11:06
[SPEAKER_00]: nowhere in this liability statement or the stipulation of liability is there an element of blame wordiness.
11:13
[SPEAKER_00]: So I'm not saying that X is blame wordy for Omega and therefore X ought to do subtly about it.
11:19
[SPEAKER_00]: Sure, there might be a subset of liability, say moral liability that says X is morally responsible for Omega, X ought to redress the effects of Omega.
11:27
[SPEAKER_00]: So if I'm an intentional murderer, I stab someone who doesn't want to kill that person dies, I'm clearly morally responsible.
11:32
[SPEAKER_00]: So what we're looking at here is a great area between the two ends of this spectrum and one end you've got the sort of clearly defined moral liability on the other hand or the other end rather you've got natural events and not caused by artificial causes and for which no one at all is to bear any liability.
11:48
[SPEAKER_00]: What I'm intrigued by is the middle ground.
11:50
[SPEAKER_00]: the questions of or the domain of liability without delay.
11:55
[SPEAKER_00]: And this is a starting point of my account where draw heavily upon the works of one wonderful political theorist of the name Iris Marin Yang.
12:02
[SPEAKER_00]: And she's truly rock writer, rather a remarkable thinker in so many different ways.
12:08
[SPEAKER_00]: And of course she was at U-Shig before she passed away very tragically around 10 to 15 years ago.
12:13
[SPEAKER_00]: And young essentially calling this framework called the social connection model of responsibility towards injustice and responsibility towards justice rather in response to injustice and essential crux of her thesis is as follows.
12:28
[SPEAKER_00]: Firstly, we as human agents are socially connected to one another provide always satisfy a number of in certain conditions that she never really quite made as clear as she could have perhaps.
12:39
[SPEAKER_00]: But she probably wanted to, had she been given more time to do so, and premise to invert your social connections to one another and more specifically the injustices of the occur on a large scale, on a collective scale, which she takes as essentially a large scale domination and subjugation by one group of individuals of another group of individuals, whereby these groups may not be behaving with intention in carrying up such domination and subjugation, and yet do so in practice and finally, from which, of course, one of these groups benefits,
13:08
[SPEAKER_00]: any other group suffers, that in these instances of structural injustices, there ought to be forward-looking responsibilities to be undertaken and born and undertaken and discharged by individuals who are connected to such injustices that do not involve blame unto themselves.
13:26
[SPEAKER_00]: So that is essentially her central claim concerning the SCM model, the social connection model of responsibility towards justice.
13:33
[SPEAKER_00]: Now very quickly, applying this to AI.
13:34
[SPEAKER_01]: My argument is, sorry, wait, but before we apply to AI, let's just, let me just see if I can simplify it a little bit.
13:41
[SPEAKER_01]: I'm going to give you a sort of a kind of toy case.
13:44
[SPEAKER_01]: It's a real story, but tell me if I'm beginning to get out and tell me what's not capturing youngs of you.
13:51
[SPEAKER_01]: I was not a particular gracious teenager, okay, so I recall, so I grew up upper middle class and Connecticut, sub but nice, beautiful suburb, whatever, lovely, bit boring, but okay, that's fine.
14:05
[SPEAKER_01]: And my parents, I forgot, we were in the backyard.
14:10
[SPEAKER_01]: My grandparents were there too, and my mother wanted me to do something, it may have been yard work or it was something that I didn't want to do that I had to do was something like general upkeep of the house.
14:20
[SPEAKER_01]: And I think my attitude was something like, I didn't agree to do any of this shit.
14:24
[SPEAKER_01]: I didn't agree to, to whatever it was, rate the leaves or spread the mulch or whatever it was.
14:30
[SPEAKER_01]: And my grandmother said, but you use the advantages that upkeep of this house.
14:36
[SPEAKER_01]: Grants, you live in the house.
14:37
[SPEAKER_01]: It's nice.
14:38
[SPEAKER_01]: It's clean.
14:39
[SPEAKER_01]: You enjoy the backyard.
14:40
[SPEAKER_01]: It's not a mess.
14:41
[SPEAKER_01]: There's even a pool back there.
14:43
[SPEAKER_01]: And so you've taken some part in Enjoying the benefits and you thereby owe your parents to help them with the upkeep of the house.
14:54
[SPEAKER_01]: Yeah, blah, blah, blah, blah.
14:55
[SPEAKER_01]: I know that there's some disadvantages if you don't know what you're talking about, but maybe you can say,
15:00
[SPEAKER_01]: Something about that situation is that the kind of thing that young is talking about or is that sort of in the neighborhood or is it a separate thing that you so think about you see of that example By the way, I was convinced my grandmother convinced me and I helped that.
15:10
[SPEAKER_01]: I thought okay shit.
15:11
[SPEAKER_01]: That was a good argument.
15:12
[SPEAKER_00]: All right, I'm not as usually extremely extremely persuasive
15:16
[SPEAKER_00]: No, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no, no
15:32
[SPEAKER_00]: which is to say that basically the example you supplied the read, the three ways of interpreting what's going on.
15:38
[SPEAKER_00]: And the first and the most obvious candidate to fix a nation is the beneficiary pays principle.
15:42
[SPEAKER_00]: And there, as advanced by, say, Daniel, the argument goes that if one benefits from saying injustice, but in your case, of course, there was no injustice, would say, if I were to benefit from an injustice, then I ought to
15:53
[SPEAKER_00]: take up the responsibility of redressing the effects of the injustice or else I'd quote and quote to your Murphy which is again someone that but quote in his works right long story short I come across as a mole slide so it's a bit like a slightly piece of work so that's one way of casting out the explanation there another interpretation is complistic so you talked about
16:11
[SPEAKER_00]: you're being complicit to some extent in using and maybe contributing towards the weeds, the mess, I don't know what the teleport term's eyes pose, the uneven and also, slavently, appearance, attempts, appearance of the yard.
16:25
[SPEAKER_00]: So therefore you're complicit in the wear and tear.
16:28
[SPEAKER_00]: So that's another way of reading the situation.
16:30
[SPEAKER_00]: The third interpretation though is the social connection model.
16:33
[SPEAKER_00]: And of course, it's worth professing that young herself doesn't quite
16:36
[SPEAKER_00]: explain how the SCM is different from or it might actually be reducible into complistity or beneficiary pay.
16:44
[SPEAKER_00]: But what he does say is that whatever it is, whatever the ground exactly is, the hype of responsibility is different.
16:50
[SPEAKER_00]: So if you're morally complistant, you might have more responsibilities.
16:53
[SPEAKER_00]: If you're benefiting from injustice, you might also cultivate just some extent a sort of moral, or at least quasi more liability, these are these what you're benefiting from.
17:02
[SPEAKER_00]: But in the event that you are connected to an injustice and merely connected,
17:06
[SPEAKER_00]: The response was that generates need not be and does not have to be blamed-centered.
17:10
[SPEAKER_00]: So then the question becomes, what does connection mean reading an ask me that question, I'm going to ask myself that question.
17:15
[SPEAKER_00]: And here's where I find me Macoens work, quite illuminating on this front where she says, actually, there's several ways of interpreting what connection means.
17:23
[SPEAKER_00]: One way is to say, okay, existence, if exist in the same society, or within the same collective, so speak more precisely, as the unjust or the perpetrators of the injustice, and all the victims of the injustice, then that existence alone counts as a connection.
17:38
[SPEAKER_00]: But she rejects, actually, that's way too broad, because he basically includes into the mix here.
17:43
[SPEAKER_00]: All sorts of,
17:44
[SPEAKER_00]: non-mull agents who exist in a society and community and not actually meaningfully socially connected and we can't actually connection to so to speak.
17:52
[SPEAKER_00]: So another interpretation, the second interpretation is dependence.
17:56
[SPEAKER_00]: In other words, if X is dependent upon a series of structural processes that generation justices, then X should be deemed as connected to these injustices.
18:05
[SPEAKER_00]: But here is where, again, we run into fairly counter-intuitive implications, which is to say that we only approve social connection in gendered responsibilities, if we're victims of some extent, on the same structures that produce season justices, right?
18:19
[SPEAKER_00]: So if I'm a sweatshop worker, I'd have to take up a responsibility to deal with global sweatshop labor, because I am a dependent upon these unjust structures.
18:27
[SPEAKER_00]: That again seems to be rather
18:29
[SPEAKER_00]: or any counter-intuitive, and that is where Macauin essentially introduces a new interpretation.
18:34
[SPEAKER_00]: She argues that social connection is best conceptualized as the reproduction of the background conditions for unjust processes.
18:43
[SPEAKER_00]: Let's unpack that in more laid person terms.
18:46
[SPEAKER_00]: It is to say that when I intentionally partake in certain activities,
18:50
[SPEAKER_00]: And by intentionally, I don't mean intending to cause harm or do harm or contribute towards the denegative, the detrimental states, but instead, I'd behave with intention and a degree of autonomy when I carry out certain actions and these actions speed into the background conditions that eventually down the line calls or precipitate, or at least in some sense create the fermenting grounds for the injustices to occur, even if there's no strict sort of counterfactual based causality to be established there.
19:18
[SPEAKER_00]: And that is what she turns to be essentially a reproduction of the background conditions for injustice.
19:23
[SPEAKER_00]: And I think that's a fairly persuasive way of reading the social connection model that isn't reducible into either complicity sent to the counts or beneficiary pays accounts, right?
19:33
[SPEAKER_00]: They might be related to the SCM, but they're not the SCM, and I think it's imperative to disentangle them.
19:38
[SPEAKER_00]: So to speak.
19:39
[SPEAKER_01]: Okay, okay.
19:40
[SPEAKER_01]: Huh.
19:42
[SPEAKER_01]: Let's change my family example for a second, which is something that I don't not for real.
19:47
[SPEAKER_01]: Suppose that my parents actually slightly murdered the previous occupants of the house, except for the children.
19:56
[SPEAKER_01]: There was a family there, it killed those parents.
20:00
[SPEAKER_01]: Those two kids became orphans.
20:03
[SPEAKER_01]: And then we moved in, I didn't know this.
20:05
[SPEAKER_01]: I was a wee lad.
20:06
[SPEAKER_01]: So it injustice has been done clearly.
20:09
[SPEAKER_01]: Compensation reparations, et cetera, are owed to those two orphans.
20:15
[SPEAKER_01]: things continue.
20:17
[SPEAKER_01]: And I guess, let's say I find out about this.
20:19
[SPEAKER_01]: I think it's because you're there.
20:20
[SPEAKER_01]: I grow up.
20:21
[SPEAKER_01]: I'm 18, whatever.
20:22
[SPEAKER_01]: I find out about this.
20:23
[SPEAKER_01]: Yes.
20:24
[SPEAKER_01]: And I can do one of two things.
20:27
[SPEAKER_01]: I can say, mom, dad, you're morally gross.
20:29
[SPEAKER_01]: This is reprehensible.
20:30
[SPEAKER_01]: I have no part of this.
20:32
[SPEAKER_01]: I am moving out tomorrow or I'm moving out with this minute.
20:35
[SPEAKER_01]: I'm going to be contacting the orphanage.
20:36
[SPEAKER_01]: I'm going to try to find these people.
20:37
[SPEAKER_01]: I'm going to try to make compensation.
20:39
[SPEAKER_01]: I'm calling the police blah, blah, blah.
20:41
[SPEAKER_01]: In that case, I don't have any
20:43
[SPEAKER_01]: I'm not responsible for having justice, I'm not playing a role in the upkeep of the injustice, et cetera.
20:49
[SPEAKER_01]: And so I don't have any sort of special moral obligations to those orphans or something.
20:53
[SPEAKER_01]: I might go above and beyond the call of duty, something I want to go with lines, but that's a certain thing.
20:57
[SPEAKER_01]: On the other hand, if I said,
21:00
[SPEAKER_01]: Mom and dad, that's some crazy shit.
21:02
[SPEAKER_01]: But yeah, your secret's safe with me.
21:04
[SPEAKER_01]: Love this house.
21:05
[SPEAKER_01]: I'm not going anywhere.
21:07
[SPEAKER_01]: I'll keep it a secret.
21:08
[SPEAKER_01]: If anyone knocks the door asking about a checkered past, I'll make sure to lie and make sure that there's deceiving we don't get found out.
21:15
[SPEAKER_01]: And so now that's me playing an active role in perpetuating the injustice, something along those lines.
21:22
[SPEAKER_01]: And in that case, I've now taken on some responsibility for
21:27
[SPEAKER_00]: I'm now properly in the realm of moral wrongs and I'm accountable for the accountability deficit that those orphans are free but I want to push back against that differentiation or that cut that you're proposing there and that's the same even if you chose to walk away so you know if you chose to walk away that's called option one now option one
21:47
[SPEAKER_00]: as you choose to walk away and report your parents to the police, right?
21:50
[SPEAKER_00]: And that's what you were hinting at.
21:52
[SPEAKER_00]: And in that case, of course, you'd be, I'd say not just doing what you're supposed to do required to do in a sort of obligatory sense, but also going up out of and beyond, so you're the super-rogatory, right?
22:02
[SPEAKER_00]: But then there's also option one B, but you walk away, you say, one none of this, I don't want to live, I don't want to live in this house.
22:07
[SPEAKER_00]: It gives me terrible amount of time.
22:08
[SPEAKER_00]: So I would tell you one about it.
22:10
[SPEAKER_00]: But, in that case, you're still actively covering up the injustice.
22:13
[SPEAKER_00]: The only difference here is that you're not emotionally attached to, or intimately associated with your parents, quay a member of the same family as you might be in the second scenario, where, of course, there's also the subject of identification, where you identify with the collective, that can be deemed as continuous over time, right?
22:31
[SPEAKER_00]: So, to be like,
22:32
[SPEAKER_00]: My saying, I know that my nation committed and perpetrated genocide had pathetically against some other nations in the past, but I'm proud of being a part of this nation.
22:41
[SPEAKER_00]: Therefore, in virtue of my subject identification, I could be held responsible.
22:45
[SPEAKER_00]: And that is a work ground-standard nationalist account of responsibility.
22:49
[SPEAKER_00]: If you look at Arash Ebersaday's writings on shame and pride, and also David Miller's writings as well.
22:54
[SPEAKER_00]: I think that's where this kicks in.
22:55
[SPEAKER_00]: But if I say, I don't like this nation.
22:56
[SPEAKER_00]: I want to walk away from this nation, but I'm not going to tell anyone about
23:00
[SPEAKER_00]: I know, which is the infrastructure season in justice as a nation historically committed.
23:03
[SPEAKER_00]: I'm still complicit in the perpetration of injustices towards the descendants of the original victims.
23:10
[SPEAKER_00]: And you can catch this out in several ways that had I actually reported on the injustice.
23:15
[SPEAKER_00]: These victims descendant would have received.
23:17
[SPEAKER_00]: compensation so that's one thing and another angle here is that my continued inability to settle the moral score in adversely perpetrics a system or sort of a relational inequality between myself and these victims to send in and that again is a second direct harm or sort of second order harm so to speak building one existing injustices and all of this is interesting fascinating and if we were to bring this back to AI because I'm sure so those listening might be thinking what is
23:47
[SPEAKER_00]: Without naming this, there's been a large number of AI more precisely data labelling, data sorting and classification companies that have drawn heavily upon exploitational wages and exploit to cheat, don't cheat Labour from global self-contrists.
24:04
[SPEAKER_00]: In Venezuela, in Kenya, and even parts of India and Bangladesh, they've drawn upon such Labour to do the quote unquote
24:16
[SPEAKER_00]: sorting through videos that might feature really obscene content and vial transmissions and trying to figure out which of these is a quote, co-show which of these is less co-show which of these is totally off limits.
24:27
[SPEAKER_00]: This is excruciating and yet the wages that these workers have received in a past
24:32
[SPEAKER_00]: are frankly not commendable with, of course, the kind of pains they had to ensure.
24:36
[SPEAKER_00]: Not even if we were to take into account purchasing power parity and, of course, you know, what a living wage constitutes in their respective constituents.
24:45
[SPEAKER_00]: Now, in these cases, I'm positive that we are all of us here using
24:49
[SPEAKER_00]: LM's using models, whatever we're using, as long as the models draw upon and supervise or supervise learning, data that's been previously labelled and of course largely supervised again somewhat reinforcement in this particular surface is to the extent we're benefiting from such models, I would posit that we therefore grew beneficiary pays.
25:08
[SPEAKER_00]: Scented or grounded liability to redress the effects of said injustices to seek to intervene and of course the jallings that we would get is how is this injustice surely sweatshop labor's not in justice Polypaid workers that's not in injustice.
25:22
[SPEAKER_00]: They could be paid more sure, but had it not been for this job.
25:24
[SPEAKER_00]: They would have been jobless with no pay whatsoever But a trouble with this theory is they think in ignores that they're certain right-space claims they're not susceptible to this kind of factual base rejoin that and that's to say that even if you would have been
25:37
[SPEAKER_00]: worse off without a certain form of intervention.
25:40
[SPEAKER_00]: The fact that this intervention occurs may be without your consent or an indignifying and disrespectful manner or in a way that frankly is less than what you deserve in an absolute intrinsic sense.
25:50
[SPEAKER_00]: That is unto itself in intrinsic and injustice.
25:52
[SPEAKER_00]: And this is where, you know, David Benetar's differentiation between intrinsic and extrinsic costs and intrinsic and extrinsic injustice is.
26:00
[SPEAKER_00]: of individuals, undergoing certain detrimental states, comes into play here.
26:05
[SPEAKER_00]: I think we can grant that these workers would have been much worse off, in some cases, maybe, had it not been given this dire labeling job, which is contingent, casualised, poorly unionised, and there's a death of protection, sure, but it's still better off working in them, but still.
26:20
[SPEAKER_00]: Better off in a welfare sense does not translate to therefore all things can sit at justified or indeed morally appropriate and normatively apt.
26:27
[SPEAKER_00]: And that is where I would say we identify cases of injustices.
26:31
[SPEAKER_00]: But this is a somewhat separate from my SCM claim.
26:34
[SPEAKER_00]: I just thought it's interesting to pick up on what you raised there on the garden and back yard example and expand it and link it back to AI, so to speak.
26:41
[SPEAKER_00]: Thank you.
26:42
[SPEAKER_01]: Yeah, they know that that is helpful.
26:43
[SPEAKER_01]: So,
26:45
[SPEAKER_01]: to do the translation, make it explicit.
26:47
[SPEAKER_01]: The house is an LLM.
26:49
[SPEAKER_01]: My parents is a foundation model company for instance, like an open AI or an open AI good.
26:55
[SPEAKER_00]: We're not saying, yeah, very good.
26:57
[SPEAKER_00]: I got excellent, excellent, I think.
26:58
[SPEAKER_00]: That's very important.
26:59
[SPEAKER_01]: A kid to attend.
27:00
[SPEAKER_01]: Then me enjoying the benefits of the house slash LLM is I'm the user of an LLM.
27:05
[SPEAKER_01]: I use Cloud, I use CHPT or whatever.
27:08
[SPEAKER_01]: Or something like that.
27:08
[SPEAKER_01]: And the orphans, of course, are, those are translated to the members of third world countries that are working, have really shitty, almost, you've basically working conditions in which the work requires a kind of suffering, a mental trauma that is not commensurate with the pay they receive, and perhaps not with any level of pay that they would receive.
27:31
[SPEAKER_00]: Of course, the orphans aren't paid to be orphans, so there's a slow dysanalogy there, but yeah, things break down, yeah, yeah.
27:36
[SPEAKER_01]: you know that that is helpful and then you ran through with the sort of obvious objection which is what they didn't have that job that they would have no job in the family would starve and the point is something like even if that's true that doesn't make it okay actually I was the case I actually gave this kind of case we were talking about the example that I gave was up people this was the context of talking about mitigating the biases of a model and it was something like but it's less biased than the alternative let's say humans
28:06
[SPEAKER_01]: And I said, yeah, just because it's a moral improvement doesn't mean it's morally permissible.
28:10
[SPEAKER_01]: So the example I gave was supposed actually gave an orphanage example.
28:13
[SPEAKER_01]: I don't know I don't know why I'm so fixated on orphans.
28:16
[SPEAKER_01]: Maybe it's because of Batman.
28:17
[SPEAKER_01]: I don't know.
28:18
[SPEAKER_01]: Not Superman.
28:19
[SPEAKER_00]: Superman.
28:20
[SPEAKER_01]: Superman.
28:20
[SPEAKER_01]: Right.
28:20
[SPEAKER_01]: Oh, that's right.
28:21
[SPEAKER_01]: He wasn't over.
28:22
[SPEAKER_00]: He was never in an orphanage.
28:23
[SPEAKER_01]: That's what crucial.
28:23
[SPEAKER_01]: Actually, now there was Bruce Wayne.
28:25
[SPEAKER_01]: He was like, how eight.
28:26
[SPEAKER_00]: Exactly.
28:27
[SPEAKER_01]: That I didn't realize they were both orphans.
28:28
[SPEAKER_01]: That's kind of a meaningful in some way.
28:30
[SPEAKER_01]: I said, look, suppose you read an orphanage, and you killed, you, or something like an at the orphanage, they got beaten every day.
28:39
[SPEAKER_01]: And then you adopt the from this brutal orphanage, and you beat them every other day.
28:45
[SPEAKER_01]: And you say, look,
28:46
[SPEAKER_01]: they're better off than they were before.
28:48
[SPEAKER_01]: They're in a morally better circumstance.
28:50
[SPEAKER_01]: They're were before.
28:50
[SPEAKER_01]: They're only getting the in every other day, not every other day.
28:53
[SPEAKER_01]: Yeah.
28:53
[SPEAKER_01]: So what I'm in the ethical clear, right?
28:56
[SPEAKER_01]: And the answer is no, you're either at the clear.
28:58
[SPEAKER_01]: Yes, you've improved their moral situation, but you're still doing something that's morally reprehensible, still morally and acceptable.
29:03
[SPEAKER_01]: You're still beating them over at the base.
29:05
[SPEAKER_01]: The fact that you morally improve the situation doesn't mean that you're in the ethical clear.
29:09
[SPEAKER_00]: And I could say that that's two observations.
29:11
[SPEAKER_00]: The first is a quick point commenting in this case.
29:14
[SPEAKER_00]: And I can actually strengthen your argument even further by noting that even into in a counterfactual, even if these orphans would have still been beaten every day in a counterfactual, right?
29:24
[SPEAKER_00]: This is still, I can't say.
29:24
[SPEAKER_00]: factual based offential is not alter the normative or more precise in the moral impromisability of your beating the children.
29:32
[SPEAKER_00]: Although of course, the pushback to you would then say, no, because you're picking the wrong amount of factors here, reading and Brian, the actual counterfactual, the moment these orphans end up in your care, right?
29:41
[SPEAKER_00]: The actual counterfactual is a world where you don't beat these orphans at all, and therefore they are worth suffering otherwise worth being, and you're not chosen to beat them.
29:48
[SPEAKER_00]: But then that's where we get to complex debate over what are the relevant counterfactuals in assessing harm and benefit.
29:54
[SPEAKER_00]: I could have been Bill Gates in some, in some sense, right?
29:57
[SPEAKER_00]: I would be Bill Gates in a possible world.
29:58
[SPEAKER_00]: That's 10,000 light years and brackets of spatial temporally disconnected from this world.
30:03
[SPEAKER_00]: Why is that really there?
30:04
[SPEAKER_00]: Could be Hillary Clinton or Hillary Clinton could have been the president in lieu of Donald Trump, whatever But that does not mean that these kind of factors are Jamae.
30:11
[SPEAKER_00]: So yes, that's what we get into what's Malansky terms in a very complex, a kind of actual stipulation or a kind of actual quave mark, okay?
30:19
[SPEAKER_00]: So back to the what were the bigger point I want to make here though, which is essentially on structural injustice a two core features that I've I personally identified in most
30:29
[SPEAKER_00]: examples of structural injustices, young or Macauin, and even robbing them, they write on, so to speak, over the past decade or so that these theories have worked on or written on.
30:38
[SPEAKER_00]: And the first feature is what we call collective harm, okay?
30:43
[SPEAKER_00]: So they are usually harm's arise, but only upon the reaching or the satisfaction of a particular threshold of scale and size.
30:51
[SPEAKER_00]: For instance, if I were to insult you, right, with a very flippant insult,
30:57
[SPEAKER_00]: on a one-on-one context.
30:58
[SPEAKER_00]: That doesn't really mean much, right?
31:00
[SPEAKER_00]: Sure, it might hurt, but it doesn't have the kind of gravitas or harm that it would be inflicted upon you if I were to say a part of a much bigger collective or a loosely defined group, right?
31:10
[SPEAKER_00]: Without coordination but still a group nonetheless,
31:13
[SPEAKER_00]: hurling these slurs at you and slurs of the bees at you and if you were of course a member of an identity group that's targeted by said slur.
31:21
[SPEAKER_00]: And I think the same analogy could be applied to say animal consumption, about eight one chicken that's already cooked on the table, that's not going to be I'm not going to be contributing towards any harm because chances are the chicken putchers or whoever is mass factory farming these chickens would produce exactly the same number of chicken.
31:37
[SPEAKER_00]: cooked chicken meats, whatever the equivalent of that is, right?
31:40
[SPEAKER_00]: Same amount, right?
31:41
[SPEAKER_00]: So no change and no delta there.
31:43
[SPEAKER_00]: But alternatively, if a thousand folks in a certain suburb town said, I'm not going to eat any more chicken, or any more chicken meat, okay, then that's going to put an end, really, to at very least, to supply his business to that very town, right?
31:56
[SPEAKER_00]: Because it's supplies me to say overheads are too costly.
31:59
[SPEAKER_00]: It's not worth it.
31:59
[SPEAKER_00]: I'm not going to deliver fresh chicken to this town.
32:02
[SPEAKER_00]: So therefore, the first point to note is that there are collective harms inflicted upon A, the chickens, and B, the identity group that's being targeted here, but only upon a satisfaction of a certain threshold of scale and intensity by extension, or separately as well.
32:16
[SPEAKER_00]: The second observation is that the actions of individuals in causing or contributing towards these injustices are contextually excusable.
32:24
[SPEAKER_00]: And here I am going to draw upon an AI case.
32:26
[SPEAKER_00]: If I am an engineer, I am an engineer.
32:28
[SPEAKER_00]: Let's say I'm installing it as an engineer, working for one of these large generated AI companies.
32:33
[SPEAKER_00]: And my job is very simple, I just need to crack this particular hog in the Outrhythmic Machine to figure out how exactly I can streamline and optimize this part of the parameters, and to reduce the time required and a complete required to get to the exact same outcome.
32:48
[SPEAKER_00]: So I'm a part of this huge sprawling machine, and I of course signed up for this company knowing that this is broadly speaking what I'm going to be doing, but I don't know who's going to use AI models?
32:57
[SPEAKER_00]: I don't know who my clients are because I'm not privy to such classified information.
33:01
[SPEAKER_00]: I'm not even a, I don't have any vested shares, right?
33:04
[SPEAKER_00]: So I'm very low down on a pecking order I'm just in an issue or fresh grad, okay?
33:08
[SPEAKER_00]: So join the company.
33:09
[SPEAKER_00]: I do my bit, and as it turns out, the model that we produce is sold to clients, including but not limited to a corrupt mafia boss operating in the Western Italy and be a warlord and in the Central African Republic using the Gen AI to propagate misinformation about children and women and also in religious minorities.
33:29
[SPEAKER_00]: So clearly, I'm contextually excusable, in a sense, I could not have
33:33
[SPEAKER_00]: reasonably been expected to know all that had I read a few more books talking about the evils of companies akin to, so the other companies you mentioned just then maybe I'd be a tad more cognizant and alert to such times, but by large, I, ignorant, I have broadly excused what ignorance about the effects of my actions and I'm really contributing that kind of actually unclear, Delta unclear amount.
33:54
[SPEAKER_00]: towards the overall injustice.
33:57
[SPEAKER_00]: Now in that case, there's contextual exclusivity and it because I am partaking in a collective harm of significance and because I am acting with intention when I'm coding and programming and the intention to complete that task where I'm not being held at gun point, no one's pointing a gun at being saying you have to do it.
34:12
[SPEAKER_00]: I voluntarily entrances contract, therefore I grew in virtue of my connection to the structural injustice there, forward-looking responsibilities, to mitigate and redress and tackle the effects of the injustice in which I am implicated in the background conditions of and reproduce the background conditions of.
34:29
[SPEAKER_00]: So that is essentially the sort of motivating example if you will to ground what I'm trying to advocate.
34:35
[SPEAKER_00]: And of course, those of you listening back at home and you might raise the following objection, which I'm going to spell out very quickly.
34:40
[SPEAKER_00]: Surely, yeah,
34:41
[SPEAKER_01]: Let me show you a little bit of a little bit of a little bit of a little bit of a little bit of a little bit of a little bit of a little bit of a little bit of a little bit of a little bit of a little bit of a little bit of a little bit of a little bit of a little bit of a little bit of a little bit of a little bit of a little bit of a little bit of a little bit of a little bit of a little bit of a little bit of a little bit of a little bit of a little bit of a little bit of a little bit of a little bit of a little bit of a little bit of a little bit of a little bit of a little bit of a little bit of a little bit of a little bit of a little bit of a little bit of a little bit of a little bit of a little bit of a little bit of a little bit of a little bit of a little bit of a little bit of a little bit of a little bit of a little bit of a little bit of a little bit of a little bit
35:09
[SPEAKER_01]: You were going to raise it up Jackson.
35:10
[SPEAKER_01]: I'm going to raise two other kinds of objections or I want to hear what I say about this and then maybe it's going to hit on them.
35:15
[SPEAKER_01]: One thing that I want to think about is causal distance or something one of those lines.
35:19
[SPEAKER_01]: And then the other thing that I want to think of is what it means to be a member of the set of people who cause the harm and how we attribute responsibilities.
35:27
[SPEAKER_01]: Okay.
35:28
[SPEAKER_01]: As of the first one, I think that we typically think that the further you, the further you removed causally from the outcome, that's really bad, the less responsible you are.
35:39
[SPEAKER_01]: There's different things to say about maybe your far back in the causal chain, but your power was that size.
35:45
[SPEAKER_01]: So, for the sake of argument, more or less assume that everyone is contributing roughly the same amount.
35:51
[SPEAKER_01]: Good.
35:52
[SPEAKER_01]: So, we're all contributing X units into the causal chain of whatever influence.
35:57
[SPEAKER_01]: We all have five units of influence how we contribute.
36:00
[SPEAKER_01]: And let's say the causal chain is I'm going to say we just pick a random number.
36:04
[SPEAKER_01]: It's a thousand links in that causal chain.
36:07
[SPEAKER_01]: And I'm linked number 23 something like that.
36:12
[SPEAKER_01]: And then this bad thing happens in 1000.
36:15
[SPEAKER_01]: Intuitively, it looks like I'm barely response for the outcome of that puzzle chain one one and there's lots of reasons one my caught my causal contribution is relatively miniscule compared to the other 999 links in the chain
36:34
[SPEAKER_01]: I couldn't have known how all is would play out because there's so many variables.
36:38
[SPEAKER_01]: There's just so many variables.
36:40
[SPEAKER_01]: I'm a finite human.
36:42
[SPEAKER_01]: I can't know all the facts.
36:44
[SPEAKER_01]: I can't know all the empirical facts on the ground.
36:46
[SPEAKER_01]: Things are confusing to me.
36:48
[SPEAKER_01]: There's other kinds of mitigating things.
36:50
[SPEAKER_01]: Like, I have to get some kind of job.
36:53
[SPEAKER_01]: I have a family obligations to my family to support my children to send them to school and give them clothing, books, and blah, blah, blah, blah.
36:59
[SPEAKER_01]: So that's a causal distance.
37:02
[UNKNOWN]: Yeah.
37:02
[SPEAKER_01]: thing that i guess i take it i know that you don't find that persuasive so i want to know why you don't find that persuasive and the second thing that i want to register before i lose track of it is
37:11
[SPEAKER_01]: It's one thing to say that the set of people is responsible.
37:17
[SPEAKER_01]: It's another thing to say that each individual member of the set is responsible.
37:21
[SPEAKER_01]: One, you can say, what's true of the, what's true of the whole?
37:23
[SPEAKER_01]: Does it mean that it's true of that?
37:24
[SPEAKER_01]: Part of the whole for those people at home who are not philosophers to take a sort of mundane example.
37:29
[SPEAKER_01]: I could say that the, whatever, the ship has the property of weighing one ton.
37:35
[SPEAKER_01]: It's not the case that every plank of wood at constitutes that ship weighs one million, what did I say, a ton?
37:41
[SPEAKER_00]: time in total although I guess in that case it would actually amount to a time in total unless you say that people want to shit them.
37:45
[SPEAKER_01]: Well there's a property had by the whole that's not a property had by the whole.
37:49
[SPEAKER_00]: But they could all have the property of being a part of something that when assembled would have one time.
37:55
[SPEAKER_00]: So there's a correlated.
37:56
[SPEAKER_00]: Yeah sure.
37:57
[SPEAKER_00]: That's fantastic.
37:58
[SPEAKER_01]: Yeah, that's fine.
37:59
[SPEAKER_01]: But I think that you talked about the meat example, like the environmental example, I throw a single piece of plastic into the ocean.
38:08
[SPEAKER_01]: It's not going to cause climate change.
38:10
[SPEAKER_01]: The problem is that the set of people, or even, sorry, driving the car, that's a better example.
38:14
[SPEAKER_01]: Driving the car is not going to cause climate change.
38:16
[SPEAKER_01]: The problem is that the set of people driving cars causes climate change, let's just simply be like for the sake of argument, but and my causal contribution is so ministerial, it's so point zero zero zero zero zero zero zero zero zero.
38:26
[SPEAKER_01]: Or when maybe it's going to reduce the causal complexity, the causal substance thing, but I'm trying to make it slightly different where it's something like, yes, I'm a member of the set of the whole that was the harm.
38:35
[SPEAKER_01]: But that has to be in that each individual member is responsible.
38:38
[SPEAKER_00]: Anyway, so I'll show that you get very good read.
38:40
[SPEAKER_00]: And I would actually stop by saying, this is exactly why young came up with a social connection model because she wants to, and I think we're co-in quite successfully completed a task envision of disentending questions of liability.
38:52
[SPEAKER_00]: or rather what she terms responsibility without liability because in her language liability means moral responsibility and that's not what she's looking for, let's just speak as priority.
39:01
[SPEAKER_00]: She wants to disentangle that from causality.
39:04
[SPEAKER_00]: She wants to disentangle that from moral responsibility.
39:06
[SPEAKER_00]: So on a causal distance claim, this is exactly why it would make no sense to say, oh, I'm implicated in this step, or I cause, okay, step 39, and therefore I am morally responsible to step 100 or step 50, even step 45, it seems pretty far fish.
39:21
[SPEAKER_00]: because, again, my ignorance is excusable.
39:24
[SPEAKER_00]: The link between 39 and 45 is too tenuous.
39:28
[SPEAKER_00]: And therefore, we should enjoy this connection and pass that moral responsibility.
39:32
[SPEAKER_00]: And yet, recognize, in a case of this social connection model or modified SCM essentially, what's
39:38
[SPEAKER_00]: at stake, it isn't so much causality, but reproduction of background conditions all.
39:44
[SPEAKER_00]: And less I be or less they'd be accused of off-escating this because it's still a causal question underneath, right?
39:50
[SPEAKER_00]: You unveil it and it's still a cause for all right.
39:51
[SPEAKER_00]: I would say it doesn't
39:53
[SPEAKER_00]: into strict counterfactual based analyses of causality.
39:56
[SPEAKER_00]: Even if it's over to term even if 45 would have happened either way, the fact that I was the 39 and not Bob not Sam not read, that is a low significant right as a ground for again this blameless form of responsibility.
40:10
[SPEAKER_00]: And to further buttresses like that, there's a chord as the penumbra of causal connection, and of course, and we're talking about the connection as tenuous as say one in two hundred.
40:19
[SPEAKER_00]: I don't think anyone, even the staunches SCM advocates would say one in two hundred, one ought to be held in a forward-looking sense responsible for two hundred, even they would say probably by that point it'd be a more intimate and proximate tense and 20s he can identify in a causal chain, for it's one could still be held responsible,
40:38
[SPEAKER_00]: But I don't think anyone is going to say 1 to 200 is still a valid distance because that's way too tenuous and extended a chain.
40:46
[SPEAKER_00]: And yet when it comes to 1 to 30 or 1 to 10, I think these are instances where the more responsibility I have may fail.
40:52
[SPEAKER_00]: They were very much here.
40:53
[SPEAKER_00]: It's such a cow's could fail.
40:55
[SPEAKER_00]: And yet the SCM could still pull through in grounding responsibility that is not counting
41:00
[SPEAKER_00]: and we take away the allegation that you, in virtue of being the 39, or step 39, must be held morally responsible for whatever happens at step 45.
41:09
[SPEAKER_00]: So I guess that's my answer to a good little distance challenge, which I think is interesting, but doesn't necessarily negate the validity and the forcefulness of young as accounts to speak.
41:18
[SPEAKER_01]: It's not that, so this is what I want to say.
41:22
[SPEAKER_01]: Okay.
41:23
[SPEAKER_01]: So I've presented a few, I didn't give it or even for that causal view.
41:26
[SPEAKER_01]: I'm just relying on it, it's not skin tuition pump.
41:29
[SPEAKER_01]: Oh, they're so far called for your move, but it's going very, very close.
41:33
[SPEAKER_01]: And what you've presented is a picture that's different from that one, where there is a kind of responsibility, call it for looking responsibility, failure to discharge that responsibility.
41:44
[SPEAKER_01]: Does it mean that you're deserving of blame?
41:46
[SPEAKER_01]: It's a different picture.
41:48
[SPEAKER_01]: And I guess I want to say,
41:50
[SPEAKER_01]: two things one and we talked a little bit about this which is so what where the teeth okay so I have a responsibility but failure to discharge our responsibility doesn't make me blameworthy doesn't make me legally up countable to so what's the sort of normative upset I guess it's not a normative yeah what's the normative upshot yeah so you've been say the words you're responsible in some way and I say okay fine put that down on my ledger I don't give a shit good is this just angels dancing on the bed yeah
42:16
[SPEAKER_01]: And then the second point, and maybe it's related to the forest, but I don't mean it to be depended on it, it's something like, okay, it's a different picture.
42:24
[SPEAKER_01]: That's an interesting picture.
42:25
[SPEAKER_01]: Why should I prefer that picture?
42:27
[SPEAKER_01]: Over the other picture that I thought, and great that I didn't give you a knockdown argument for it, but you own giving me a knockdown argument for the alternative picture about forward looking responsibility.
42:36
[SPEAKER_00]: I think I'm going to work backwards here.
42:38
[SPEAKER_00]: Gruntingly, in fact, some of the injustices we're talking about here are real and gruntingly, we accept that these are injustices.
42:45
[SPEAKER_00]: There's a serious moral burden, or a sort of moral remainder, so to speak, in relation to unfifilled responsibilities towards the past.
42:54
[SPEAKER_00]: And there's also, of course, correlatively, both present and also likely to be arising in the future, responsibilities towards injustices of a similar kind.
43:02
[SPEAKER_00]: So this is a practical observation that they exist.
43:05
[SPEAKER_00]: This need, this demand for response, okay?
43:08
[SPEAKER_00]: And I'm using very vague words here because I think there's a lot of ways of interpreting and casting out the demand for response.
43:14
[SPEAKER_00]: Okay, we start with that premise.
43:16
[SPEAKER_00]: Then of course the question is, what gets, right?
43:18
[SPEAKER_00]: Do we prefer a world we say, hey, I'm not morally in brackets, cause leap plus some other further stipulated conditions say, stores in or stores only in or Frankfurt, whatever you call it, these conditions, whatever you like, okay?
43:30
[SPEAKER_00]: I'm not morally responsible.
43:32
[SPEAKER_00]: Okay, this probably exists.
43:34
[SPEAKER_00]: I'm not wearing a song sport, bye-bye.
43:36
[SPEAKER_00]: Couldn't buy it all as well.
43:37
[SPEAKER_00]: I'm God.
43:37
[SPEAKER_00]: I'm out here.
43:39
[SPEAKER_00]: I see.
43:39
[SPEAKER_00]: But in a second story as I say, I'm not wearing a song sport, but I invert you of being connected to the injustice here via or query reproduces of the background conditions.
43:49
[SPEAKER_00]: And therefore, in a forward-looking sense, responsible.
43:53
[SPEAKER_00]: And I'm willing to discharge said
43:57
[SPEAKER_00]: through, for instance, raising awareness about the problem, through donating towards causes attack or these issues, or through pushing for governmental reforts and advancing institutional changes.
44:09
[SPEAKER_00]: And I think from the get-go young has made very clear that her theorization has to have practical upshots and practices.
44:16
[SPEAKER_00]: And I'll deal with the angels and the pain and what gives and teeth questions shortly.
44:20
[SPEAKER_00]: But before we get on to that, I think the reason why intuitively prefer this story is it's just enriches and allows for the fine graining of responsibilities beyond the, I'm blameworthy, I'm not blameworthy, Tada, binary and dichot, so to speak.
44:33
[SPEAKER_01]: Yeah.
44:34
[SPEAKER_01]: OK, so let me, I like that response, but let me put it back on it.
44:37
[SPEAKER_01]: So the starting point, the launching point for that,
44:41
[SPEAKER_01]: response is something like, there are these injustices that demand a response and here's another way you're putting that that must be this is my way but this is one this is a way that one might put it I don't have a strong view on this one might say it here's what I'm looking at these really unjust things happen and it sucks
45:01
[SPEAKER_01]: But that's just the way the world is.
45:03
[SPEAKER_01]: Shitty things happen.
45:05
[SPEAKER_01]: There's not a kind of moral balance and herding the universe, or even in the social universe.
45:11
[SPEAKER_01]: And yeah, it might be lovely, it might be nice if we could figure out and do the moral math.
45:17
[SPEAKER_01]: And it all comes out that all the injustices,
45:20
[SPEAKER_01]: There's a corollary set of people responsible for it and that helps me sleep at night because there's a kind of moral order that I could rest my hat on and then in full full blown niche in style I want to say something like That's a lovely comforting story that you've told yourself where there's a moral order
45:36
[SPEAKER_01]: to the universe, social or otherwise.
45:38
[SPEAKER_01]: And so you've invented some kind of forward-looking responsibility to do that moral accounting that makes you sleep better at night.
45:46
[SPEAKER_01]: But that's all it is.
45:47
[SPEAKER_01]: It's a useful fiction for trying to make the world an overly sensible place, one in fact.
45:55
[SPEAKER_00]: It's a morally non-sensical absurdity place, but doesn't that objection prove too much in terms of also eliminating the grounds?
46:03
[SPEAKER_00]: Doesn't that also eliminate grounds for backwards-oriented responsibilities?
46:07
[SPEAKER_00]: Because we might, I can pull the same card on you and say, Sure, read you might want to say it, and you killed someone.
46:13
[SPEAKER_00]: You're morally responsible for killing someone, and therefore that ought to be punished for it.
46:17
[SPEAKER_00]: But punishment.
46:18
[SPEAKER_00]: Retribution and also holding someone to account over wrongs they committed in the past.
46:24
[SPEAKER_00]: These are, this is only a useful fiction.
46:26
[SPEAKER_00]: It doesn't exist in the universe, right?
46:29
[SPEAKER_00]: This is a moral fabrication, or morality is a fabrication.
46:32
[SPEAKER_00]: In fact, I would say, the GD to rescue drowning children, and again, that's a fiction.
46:36
[SPEAKER_00]: This is a story that we tell ourselves, but in fact, we should live in a world where we are strive towards excellence.
46:42
[SPEAKER_00]: We have to live in the world to power, to define our moral condition
46:48
[SPEAKER_00]: So I think that proves too much, that's my worry, because I know this is true.
46:52
[SPEAKER_00]: Push is perhaps prove too much.
46:54
[SPEAKER_00]: And of course, the stronger version of this, by the way, is to say, hold on a second, because at least in the cases where you committed a wrong, there's a clear attribute ability.
47:02
[SPEAKER_00]: Okay, and that's when bringing an attribution that to Tersu Makka, Angela Smith, you name it.
47:07
[SPEAKER_00]: There's clear attribute ability, you are the individual who, you know, for whatever reason, right, I can allocate it.
47:13
[SPEAKER_00]: I can point to this and say, this is part of your moral ledger.
47:16
[SPEAKER_00]: because you caused it and you maintain it for it to happen.
47:19
[SPEAKER_00]: But in case of a structural injustice in gendered or grounded responsibility, it's not really actually usable.
47:24
[SPEAKER_00]: I can't say read you've developed this responsibility because you're read.
47:28
[SPEAKER_00]: This is a general duty or general responsibility that applies to read, bribe, and perhaps all of you listening back.
47:34
[SPEAKER_00]: Oh, so surely there's an asymmetry here.
47:37
[SPEAKER_00]: Now I'm not convinced to say symmetry therefore proves that one is real and the other is not.
47:41
[SPEAKER_00]: That's my first response to this challenge and the second thought is maybe this is the beauty of the social connection model and that's where we'll eventually return to the question about teeth later and how we can enforce this later but by large we don't only have actually mutable responsibilities or attribution in gendered responsibilities in our lives.
47:59
[SPEAKER_00]: the duty to donate to the duty to save drowning children.
48:02
[SPEAKER_00]: This is not because I'm read one Brian that I am there for in position of this duty.
48:07
[SPEAKER_00]: It's instead in virtue of the relationship between myself and, of course, individuals in distress that are contingently generated by the circumstances in which we interact, right?
48:16
[SPEAKER_00]: So if I were not walking past a drowning child in a shallow pond, I would not have that duty activated.
48:21
[SPEAKER_00]: And
48:21
[SPEAKER_00]: And I would not be submitting a wrong, if I'm not, I can play with that GC or at least here.
48:25
[SPEAKER_00]: I wouldn't be seeing a sub-erogatory because I'm not in a position to save.
48:28
[SPEAKER_00]: That drowning child, five miles of work.
48:30
[SPEAKER_00]: So that's just what I want to say in response to your pushback.
48:33
[SPEAKER_01]: Okay, that that that that is helpful.
48:35
[SPEAKER_01]: I'm not sure what the last part.
48:36
[SPEAKER_01]: I think that the fact that you're walking by makes it distinctly trivial to you and ask someone else because they are the person.
48:41
[SPEAKER_00]: But it's contingent.
48:42
[SPEAKER_00]: But that's right, right, right.
48:43
[SPEAKER_00]: Yeah, exactly.
48:44
[SPEAKER_00]: But put that to the side.
48:45
[SPEAKER_00]: That's what we do.
48:45
[SPEAKER_01]: Yeah.
48:46
[SPEAKER_01]: Let me ask you this and this may maybe speak.
48:48
[SPEAKER_01]: Get this back to the tea question.
48:50
[SPEAKER_01]: What's the difference between saying?
48:52
[SPEAKER_01]: It would be nice if people did some things about this injustice.
48:57
[SPEAKER_01]: And people are for looking responsible.
49:01
[SPEAKER_01]: to do things by this injustice.
49:02
[SPEAKER_01]: So there are differences between those two things.
49:03
[SPEAKER_01]: And the first it's meant to be, I hear in a super auger story sense, you're going above him.
49:07
[SPEAKER_01]: Yeah, you'd be really nice, but you're not really cool.
49:09
[SPEAKER_01]: And in the responsibility, I take it as a kind of requirement.
49:11
[SPEAKER_01]: So it's not super auger to me.
49:13
[SPEAKER_01]: Is there a, is there that difference?
49:15
[SPEAKER_01]: I think there's a, and b, if there is that difference, how do we know?
49:19
[SPEAKER_01]: So to speak that we're talking about
49:22
[SPEAKER_01]: a kind of responsibility instead of a kind of moral goodness.
49:25
[SPEAKER_00]: Very good.
49:26
[SPEAKER_01]: That's above and beyond the call of duty.
49:27
[SPEAKER_00]: Young gave us this the seeds of an answer in her account, where she talked about the political response that this really grew.
49:35
[SPEAKER_00]: And yet, of course, you know, whenever we mention political responsibility, folks get allergic, right?
49:39
[SPEAKER_00]: They start saying, oh, you're talking about political duties, political legitimacy, social contract, all of that.
49:44
[SPEAKER_00]: And then I think this is a misreading of young.
49:46
[SPEAKER_00]: Because
49:46
[SPEAKER_00]: This whole reading of the political is in some ways informed by Hannah Aratt, right?
49:50
[SPEAKER_00]: And it's informed by this concept of the demos, the collective, the political being an organic, sort of part of the fact that diva and action that we pursue as members of the human species.
50:02
[SPEAKER_00]: So setting aside all of these lofty ideas and factors that she would posit that in contrast to the general ill-be-nice of he did it in brackets in a moral sense.
50:11
[SPEAKER_00]: Specifically, one has in a political sense forward-looking responsibilities, that one can cash out through collective organization, mass assembling, protests, or agitating for change via political institutions and structure.
50:24
[SPEAKER_00]: So for her, it's a wholesale package, whereas I'm actually quite flexible.
50:28
[SPEAKER_00]: I don't think forward-looking responsibilities have to be discharged politically.
50:31
[SPEAKER_00]: I could also discharge it apolitically by corporations or individually without invoking others in my community, if I'm very powerful, if I'm under Winston Lord or Lex Luther coming to reconcile with my past wrongdoings, I might just pull this off as a one-man ban, even as a super-man, so to speak, right?
50:48
[SPEAKER_00]: So it doesn't have to be a collective discharging process, so to speak, fees and fees responsibilities, but it needs to be political, and that is what young argued
51:00
[SPEAKER_00]: to explain the conceptual distinction between the two interpretations you've raised there.
51:04
[SPEAKER_00]: But of course, then the question becomes, what does that mean in practice?
51:07
[SPEAKER_00]: And this is what I'm going to kill two birds with one stone, even if that's anti-Broadrites, which is sad.
51:13
[SPEAKER_00]: But I'm going to say that we can actually resolve both the teeth problem and the distinction problem by invoking the notion of criticisability.
51:20
[SPEAKER_00]: Again, this is a Irish-Marianne construct, which he says, one may not be blameworthy for failing to discharge duties, or forward-looking, but one can be criticised as such.
51:29
[SPEAKER_00]: And of course, there's been a lot of delay back and forth, secondary literature, Jeff Rayman, Martha Nussbaum, saying, what does this criticise ability mean?
51:37
[SPEAKER_00]: So I criticise you for all sorts of things, I criticise you for speaking to quickly and criticise you for having a very dynamic interview style, whatever, but such criticism does not amount to the same kind of criticism as what young would have expected, right?
51:51
[SPEAKER_00]: Those who repeatedly ignore other thoughtless responsibilities to have levied towards them and directed towards them.
51:58
[SPEAKER_00]: So this is where I would say this is a normative sense of criticism, one is normatively criticized for all, if one fails to discharge forward-looking responsibilities, and we're not such normative-criticisability therefore translates to a permission on a part of other actors to intervene.
52:13
[SPEAKER_00]: to require you perhaps, or at least to strongly nudge you to take up these responsibilities, that I think remains a hitherto underexplored and discussed aspect of Irish mariongians' work, which is essentially, does criticisability give rise to third party, centred and grounded responsibilities to enforce, whatever one is being criticised for, not upholding or implementing in practice.
52:38
[SPEAKER_01]: Yeah, so there's a way in which I'm with you and I wish them not.
52:41
[SPEAKER_01]: So One thing that it feels to me that you're doing is you're making criticism Quite flirtatious with I think there's a way of criticizing that doesn't is an associated with blame and there is a kind of criticism that it's
52:53
[SPEAKER_01]: So, an example that, you know, occurs to me of criticizing that's not blame, that's not also blaming.
53:00
[SPEAKER_01]: I might say that some kid is a spoiled brat.
53:03
[SPEAKER_01]: This kid's a spoiled brat.
53:05
[SPEAKER_01]: And you might say, that's a criticism.
53:07
[SPEAKER_01]: And you might say, if that's his fault, it's his parents.
53:09
[SPEAKER_01]: And I say, I'm not blaming the kid for being a spoiled brat.
53:12
[SPEAKER_01]: It's his parents, falls on him.
53:13
[SPEAKER_01]: But nonetheless, the kid's a spoiled brat.
53:15
[SPEAKER_01]: So, I could see how criticism can come apart from blame.
53:19
[SPEAKER_01]: Maybe you think about it as something like it's an evaluation of a character Independent from any evaluation of the responsibility for the character they formed.
53:26
[SPEAKER_01]: That's a cut.
53:27
[SPEAKER_01]: Anyway, so I could see that But now once we get to intervention Nudging slash forcing compelling We're really I think we're beginning to get into the realm.
53:38
[SPEAKER_00]: Yeah, so we need blame.
53:39
[SPEAKER_01]: You didn't do this thing and now I'm gonna get you to do it
53:42
[SPEAKER_01]: It's at least something quite significant because we're at least doing things like we're now trying to potentially impenetrable our autonomy in the name of getting them to discharge the responsibility So you might not call a blame, but if it looks like blame and blah, blah, blah, blah What about your nan?
53:57
[SPEAKER_00]: Was your nan blaming you when she said you've got to do your Jewish share for the backyard and a god?
54:03
[SPEAKER_00]: It probably she was playing this is only there right when she said read you've got to do this You know, that's not a sign that you're blameworthy
54:10
[SPEAKER_01]: Well, but if I hadn't, you wouldn't blame me on this.
54:12
[SPEAKER_00]: You would say, hey, why haven't you done what you're supposed to do?
54:15
[SPEAKER_00]: Go do it.
54:16
[SPEAKER_00]: And you would say, okay, I'll do it.
54:17
[SPEAKER_01]: You know, she would have blamed me for having not done it.
54:19
[SPEAKER_01]: I mean, like, she would have figured I'd done the wrong right.
54:22
[SPEAKER_00]: My focus, I made the decision to do the wrong thing.
54:25
[SPEAKER_00]: Okay, maybe in some other cases that are equally plausible they're not your man and yourself, the man would say, hey, you really ought to do it.
54:31
[SPEAKER_00]: This is a part of what you ought to do.
54:33
[SPEAKER_00]: But she wouldn't say, and not doing it, your blame worthy, or I'm in a whole, do you blame worthy for not doing that?
54:39
[SPEAKER_00]: You know, that's, I guess that this is where our intuitions come apart, which is essentially how fine-grained.
54:44
[SPEAKER_00]: And this is an interesting methodological point as well.
54:46
[SPEAKER_00]: How fine-grained to our normative intuitions and normative instincts and reactions to others, such that we can actually allow for maintenance of this, and all most, you might say, very much like angels dancing on a pinned distinction in the eyes of some, between blame-worthyness, criticisability and just a complete
55:09
[SPEAKER_00]: And I suppose this really depends on the cultural concepts and also the normative settings and behaviors of the people around us and in which we operate.
55:18
[SPEAKER_00]: So it's a very good point you're raising that.
55:21
[SPEAKER_01]: Okay, so I think at the end of the day, let me see if I could just paint the picture.
55:26
[SPEAKER_01]: The picture is one according to which we have these four looking responsibilities.
55:31
[SPEAKER_01]: If we fail to discharge those response, and we have those responsibilities by virtue of us playing a role in injustice, or specifically the role that we play is one in which we help to reinforce the reproduction of the systems that create the injustices.
55:47
[SPEAKER_01]: And for our role, we are not necessarily blameworthy in some deep and important way, but we are nonetheless potentially in breach of for looking responsibilities.
56:01
[SPEAKER_01]: Because we are in breach of forward-looking responsibilities, there's a sort of open question to be had about the extent to which people can intervene and Get us to do the thing that we ought to do even though we're not blame where they if we don't do it
56:15
[SPEAKER_01]: That's the picture.
56:16
[SPEAKER_01]: Yeah.
56:17
[SPEAKER_01]: I mean, that's super fascinating, super helpful.
56:20
[SPEAKER_01]: And I hope the art is just walking away with, we started on things like buy, stay eye and privacy violations and who's responsible, but it's very, to me, and I hope to the art is very edifying to see the way in which you couldn't start with this thread.
56:36
[SPEAKER_01]: of this is terrible shit and then back into okay now let's think about what how does that make us think about if you're like our moral universe more broadly?
56:45
[SPEAKER_00]: Absolutely and I think I just want to say read so those of you who listen back in home read is really quite an exceptional podcast host so lucid as so methodical in a way he interrogates or rather gets the best out of his guess I really would encourage you to a follow and listen to and subscribe to read because I really
57:06
[SPEAKER_01]: Yeah, so did I appreciate it.
57:07
[SPEAKER_01]: I definitely will cut that part out.
57:09
[SPEAKER_01]: And yeah, we'll have to do this again.
57:10
[SPEAKER_00]: You'll I'm sure you've got lots more views for us to say thank you read and just to fly.
57:14
[SPEAKER_00]: If I may very quickly at these views and these arguments that I discuss in this podcast, I'll be encapsulating or presenting.
57:21
[SPEAKER_00]: You can dense version of it in an upcoming book hopefully to be published no later than Q1 next year.
57:27
[SPEAKER_00]: on the ethics of geopolitics and AI or ethics of the geopolitics of AI.
57:33
[SPEAKER_00]: And that's going to be coming out next year with a C.P.
57:36
[SPEAKER_00]: So thank you guys very much.
57:38
[SPEAKER_01]: Yeah, no, that's great.
57:39
[SPEAKER_01]: We'll put that in the show notes.
57:40
[SPEAKER_01]: And when it does come out, let me know in the last time.
57:42
[SPEAKER_00]: Sure, make sure to post it on my social networks.
57:44
[SPEAKER_00]: I appreciate it.
57:45
[SPEAKER_00]: Have a lovely day.
57:46
[SPEAKER_00]: Take care.
57:46
[SPEAKER_00]: Thank you.
57:46
[SPEAKER_00]: Bye.
57:47
[SPEAKER_01]: Ciao.
57:49
[SPEAKER_01]: Thanks everyone for listening.
57:51
[SPEAKER_01]: I hope you enjoyed the episode, and I want to thank a few people who make this all possible.
57:55
[SPEAKER_01]: Delano Alexander, he's my producer.
57:56
[SPEAKER_01]: He does such an amazing job on everything that he touches.
57:59
[SPEAKER_01]: Garrett Lane is the sound engineer.
58:01
[SPEAKER_01]: That's honestly, he's just, he's too good for me.
58:03
[SPEAKER_01]: Authamia Jairam has joined us recently.
58:05
[SPEAKER_01]: He's helping us find the best guests to bring on to this podcast.
58:09
[SPEAKER_01]: And Ashley Wasselenko helps us organize all this.
58:11
[SPEAKER_01]: Get people on the schedule, get people's bios, headshots, and really keeps the train on the track.
58:16
[SPEAKER_01]: So thanks guys, really appreciate all the hard work you do.
58:29
[UNKNOWN]: Thank you.