Patreon Preview: The Fascistic Solutionism of AI w/ Dan McQuillan

0:00 Hey, this is Paris. I hope you enjoyed the Data Vampire series that we did back in October. It's had a fantastic response. And for that series, I spoke to a bunch of experts. And now we're releasing the full-length versions of those interviews for our supporters over on Patreon.com. And I wanted to give you a preview of what those interviews sound like so you can consider whether to go to patreon.com slash techwon'save us, become a supporter yourself, so you can learn even more about the important topics that we dug into in that special series.

0:30 So enjoy this clip from my interview with Dan McQuillan. I feel like on the narrative side of things, as you're talking about, right? There's this idea of, you know, on the one hand, you have these technologies that are inherently powerful and that can do so many things that we didn't think that they could do before, right?

0:46 And a lot of that is deception and not realized in the actuality of the technology. But then we also have kind of the term artificial intelligence and how it suggests a lot of different things, but also is this kind of broad overarching term for many different forms of technology that can just be deployed, you know, whichever way people think.

1:07 I wonder about what you think more about that narrative side of things and how it is deployed by these tech companies for their own benefit, really, and to try to cover over those harms that you're talking about.

1:19 I mean, I would generally refer to AI as a very, as you say, is a huge blanket term. I tend to pick on things that are based on some form of deep learning onwards, so sort of neural networks and their works, you know, which includes sometimes called discriminative AI, you know, things like image classification or recognition, and then on up to generative AI and it's you know text generation or image generation, whatever.

1:42 That's what I'd call AI, but of course AI is used to refer to all sorts of things to much more rudimentary forms of machine learning, I guess, or much more straightforward forms of machine learning, and even non-machine learning algorithms. So it's a very broad term. But I think the things that gives us its sort of epistemic power is what appears to be the accomplishments of these kind of cutting-edge models, especially now that they're accessible to everybody.

2:04 You know, before there were just sort of abstract claims in the uh, let's say in the media scape about this powerful thing called AI. And then, you know, there's the phenomena of something like self-driving cars or, you know, allegedly self-driving cars. So there are videos of you know vehicles that seem to be able to steer themselves.

2:21 You know, we all know that hasn't gone particularly well, but you know, it's still something that seems very sci-fi. That's essentially what it is, is tapping into those sci-fi agendas that we've all been sort of pre-inoculated with through various kinds of media and movies and what have you, and literature.

2:36 So it's it's a form of technology that's doing a good enough job. It's an actual technology, a concrete technology, it does real things. It connects sort of sufficiently plausibly to a sort of cultural narrative about the possibility of artificial intelligence, which already has so much projected into it.

2:53 I mean, it's you know, it's it's just a kind of contemporary version of a sort of a centuries-old sort of sci-fi literature now. And that already has so many social, psychic, and cultural imaginaries and hopes built into it. But it's also it's not really powerful. That's the kind of weird thing. This is a little bit of a, as they say, in a pet peeve of mine, but it seems that even in the space of, let's say, sociological or journalistic critique of AI, people feel almost obliged to preface their statements with, of course, AI is incredibly powerful and will definitely deliver benefits, but, and I'm just not seeing it.

3:29 I mean, AI is essentially a bullshit technology. I mean, it's it's very sophisticated and can do definitely do some computational things that I never thought were possible, but that's still not the things it claims to do. You know, the fact that it can synthesize language in this increasingly plausible way or generate images that might be entertaining or dramatic or even concerning, you know, or any of these things that, yeah, I wouldn't necessarily have said it would be able to do before, but it's also a bit like, so what?

3:55 You know, so it can do those things. That's a thing. That's not the same thing as delivering healthcare, it's not the same thing as planning economy, it's certainly not the same thing as detecting cancer. I mean, none of these things are actually done by these computational tricks. So AI is essentially, I would say, a bit of a well, at the moment, because of the amount of investment and commitment and sort of strategic sort of uh weight put on AI, I would say it kind of amounts to a sort of international Ponzi scheme, really.

4:21 If you found that clip intriguing, go to patreon.com slash techwon'save us, where you can become a supporter, support the work that goes into making the show, and get the full length version of that interview. Thanks so much.