S1: Welcome in San Diego. It's Jade Hindman on this episode of KPBS Midday Edition. A new documentary from PBS Independent Lens traces the origins of artificial intelligence to eugenics. This is KPBS Midday Edition. Connecting our communities through conversation. Over the last few weeks , we've been hearing all about fears around AI and its potential dangers for humanity. But some experts say that talking about a hypothetical AI doomsday is really a distraction from real harms we're seeing today. A new documentary from PBS , Independent Lens traces the origins of artificial intelligence and its eugenics roots. The film is called ghost in the machine.
S2: Standard deviation correlation to logit model. Logistic regression emerged specifically out of eugenics.
S3: Which started to define human cognition in machine terms.
S4: A man's brain might work as a machine.
S5: Our descendant will not be the child of the loin , but the child of the brains , the thing we call the computer. Hmm.
S1: Hmm. I'm joined now by the film's director , Valerie Veatch. Welcome to midday Edition , Valerie.
S6: Thank you for having me.
S1: So glad to have you here. Let's start with the origin story for this film. What inspired it?
S6: So probably about a year and a half ago now , one of my good friends from university signed me up for this invitation only artists working group for a , you know , secret AI company. And this was going to transform filmmaking and it was a Sora OpenAI , Sora before they had launched the product. And in interacting with this generative AI technology , so-called , Um , you know , I kind of was stunned by how the reality of this tech was a , you know , producing highly racist , highly sexist images , which is no secret. Um , but I was very astonished that this didn't seem to bother the company , and I couldn't really get any answers. Like , why is it doing this and why? Why is this , you know , a product that you can ship. And so I started to do some research. And out of that research came this film. I was really fortunate that so many wonderful experts across many fields took the time to chat with me. And out of those conversations came this film. Wow.
S1: Wow. Well , this documentary , I mean , it takes us back to the beginning to explore who built AI and why. And we quickly learn that AI is rooted in eugenics , the racist pursuit of selective breeding that emerged in the 1880s. Can you connect the dots for us between eugenics and the origins of AI.
S6: There are two threads here. One is the use of statistical algorithms in machine learning. And a lot of these algorithms , as we heard in that clip that you played , uh , come directly from the eugenic project. And I think one of the interesting elements there is that the act of producing a data point , uh , then to , you know , find similar data points is a very political act and a very loaded act. And I think sometimes we look at machine learning as this neutral set of algorithms , but that's just simply not the case. And then the other track that's quite interesting is the we hear the term especially right now in the news , um , AGI or artificial general intelligence. And this is meant to be a , you know , an AI system that becomes so good at self-learning that it can improve itself and become the smartest , you know , AI system on the planet. And that's going to be AGI. And the G in AGI is the general refers to general intelligence which is a eugenic concept. And it was this , this , this attempt to try to find a kind of measurable generalizable index for human intelligence. And the project was always to link that to race. And so what we see now in computer science , which is really fascinating , is a lot of these papers on AI , as recently as a Microsoft paper that was published in 2023 , um , refer to definitions of intelligence that are , uh , racialized and come from that , um , lineage of , uh , work. So the , the G in general of an AGI also connects to general intelligence. So it's kind of on two tracks. It's the statistics within the machine learning. And then also the , uh , the definitions we're using to define intelligence currently. Huh?
S1: Huh? Um , and I want to get back to that , but talk to me more about this movement. This where AI and race , um , connect. And what that is actually rooted in is what is this? Is this transhuman movement? What is this?
S6: Yeah , that's a very good question. I transhumanism is a huge part of this. And we didn't even get to get into this in the film. And you're absolutely right that the , um , the kind of the , the entire AI project we can call it , um , stems from wanting to create a machine that can produce the intelligence of a human. And as soon as we start to get into measuring intelligence , that's when we get into the racialized legacy of this work. And then , you know , we can look at currently a lot of the outputs of AI systems , whether it's chatbots or whether it's predictive policing. Um , all of these systems carry with them a , and amplify existing biases within data sets. And so that there's a limit to this.
S1: Have did you have you been able to get close to the why?
S6: You know , I think that there's a worldview that is permeates this kind of project , and it's this worldview of , um , optimizing measurable performance. Right. And when we focus on that , um , without understanding , you know , kind of our own position ality in , in asserting these claims , then that's where I think some of the , um , you know , supremacy comes through. But the why is difficult. You know , I think it is this kind of hyper rational , uh , worldview that is problematic as we're seeing today. But it you know , that's a very good question. I don't know if I got close to the why? Other than , you know , a hierarchical way of thinking about life is something that some people latch on to and some people don't. And I think that that's , you know , a challenge. Mhm.
S1: Mhm. You know , another big theme of this documentary is the influence of misogyny , specifically white misogyny in the AI space. Here's historian doctor Becca Lewis.
S7: There is this fixation.
S8: With women's ability to give birth , and kind of this quest for men to be able to capture that through the building of technology. Women give birth biologically , so men should be able to be the ones to give birth to new startups , new technologies. And really , this fixation on creating a patrilineal structure within Silicon Valley that doesn't need women there.
S1: So talk to me. I mean , how do the ideas of white male Superiority drive this movement.
S6: I think you hit it on the head with transhumanism , this this notion that we can improve our human stock not only through genetic breeding , which is the eugenics portion of this , but also through intervention , through , you know , implanting our brains or , you know , we kind of maxing our intelligence through technological intervention. And , you know , ultimately , I think what we're seeing now is , is by engaging with this term , artificial intelligence , which is very much , I want to be clear , a marketing term. And we see that in the film. It was created to raise money. Um , but what we see with that project is this displacement of the natural , and it's the displacement of the planet as a priority , but also this displacement of women being involved in , as we hear Becca talk about in the , you know , the progeny of life or whatever. and that is. There's a clip that I didn't put in the film that just stays with me , and it's this podcast between Sam Altman and this podcast , bro. And he says , oh , Sam Altman , the ultimate tech lord. You know you're going to have a baby. Isn't that cool? And he's like , yeah , I'm having a baby. And then they start to talk about how , you know , actually , you know , in China they're developing , uh , artificial wombs for babies to be grown in. Now , that's just a very huge stretch. That's not really what's happening. But it was this. And then this AI generated image of , like , a fetus growing inside of this , like , glass case flashes by and they're like , yeah , yeah. Well , clearly that'd be a more optimal environment to like for a baby to grow in. And I just had to stop for a moment and it's like , where are we losing our anchor that , you know , a machine is going to now grow a baby? Like , that's just impossible. You know , the hormonal interchange between , you know , a a body and a baby is profound and all of those things. And so it just seems like there's this , you know , fantasy that women can be fully displaced from this process. And part of the project of this film is to point out that these ideas that are taken as truth , kind of currently in the space we're in are , are , are all constructs. They're all kind of made up things on made up things. And now we arrive at this moment in 2026 where we're genuinely , you know , hearing one of the most , you know , powerful , wealthy men on the planet be like , yes , obviously , you know , an artificial womb is is superior. And , you know , I just I think that really we need to pause and contemplate the kind of not only the legacy of these ideas , but also the trajectory of these ideas and challenge.
S1: Which I know you've done a lot of you question where did the where's the anchor in all of this? Um , where is it? What do you think so far? Obviously , you know , if we had the million dollar answer. Problem solved. But , I mean , what's what is the deepest insecurity here that drives this?
S6: You know , there seems to be a real rejection of the body and a rejection of the natural. And this is , you know , this is not a new kind of way of thinking , but it's condensed into this moment. But the the story actually starts , you know , in my mind before the enlightenment , like , you know , when we were burning the witches in Europe and there is this kind of effort to negate knowledge spaces outside of the church or outside of political structures and , you know , pressing into the present day , you know , the only antidote to this kind of way of thinking is very boring. It's the classic kind of stuff. It's it's rooting ourselves in our communities , right? Neighborhoods are the and communities of people gathering locally to contemplate resources. And , uh , you know , our vision for our society that happens on a , on a local level. And I think that's kind of where we see in , in the US an amazing movement happening , right. You know , communities who are otherwise not inclined to become political , becoming political around the expansion of hyperscale data centers and the kind of material intrusions that those apparatuses like , require , um , are , are galvanizing local communities. And I think that's a very interesting thing to note. And , and I think that that's sort of like where we can go from here.
S1: Um , you know , I mean , this film also explores the real human toll of of what it takes to power AI in today's world and the predatory practices of these tech companies. What did you learn about what's happening in places like? I mean , there's things happening here , uh , stateside , but also in places like Kenya. Yeah.
S6: Yeah. Um , in terms of the hyperscale data center expansion specifically , there's , you know , this is happening all around the world. What is astonishing to us , as on the film team , is how many requests we're getting to screen the film. You know , obviously in America , but in Germany and France , in Sweden , and we're going to do a premiere in Nairobi. And , you know , all of these communities are dealing with the exact same issues. It ultimately comes down to , you know , municipal water supply access. You know , how we handle the power grid , how we distribute power , how we understand resource allocation around these hyperscale data centers. So that's I think one thing that's quite shocking is how global the hyperscale data center buildout is. And then therefore the community organizing is a global. It's amazing.
S1: Well , you know , it seems like this documentary is coming at a very important moment. Um , the U.N. , a U.N. committee is taking this up today as well , giving all the. Given all the recent AI news , how is the current news around AI really bringing new meaning to this story?
S6: I think what's interesting that the story arc of the last two weeks , let's say , you know , we we first we got a whistleblower from anthropic who was like , oh , these systems are going to be so powerful. We're all terrified , which we have seen before. I mean , to be clear , Dario thought that GPT two was going to be superintelligence and we should slow down. So every time these guys need to fundraise , they they put out this. These things are so scary. We must slow this down. But that narrative now rings a little bit more hollow than it maybe did two years ago. And ultimately , what the kind of current story around. Oh my gosh , these systems are so powerful. We need to slow down. We need to regulate them. Yeah we do , but let's talk about what we're regulating first of all. But then bigger than that. And this is where , you know , it's interesting to watch the story unfold. There's sort of like a well , the taxpayers need to have ownership and need to buy into the system. And let's be clear , that's a bailout. And so , you know what? What is really important to kind of declare right now is that this fantasy that machines can think is a fantasy , and it's built on a marketing term construct and any , any kind of hyperbole that these , you know , uh , like AI pioneers are throwing around is only to enrich themselves and to evade accountability for the fact that they've led a huge CapEx investment in an industry that has no return.
S1: All right. This is a very interesting conversation. I've been speaking with Valerie Beach , director of the PBS Independent Lens documentary ghost in the machine. You can watch the documentary for free on PBS or YouTube. Valerie , thank you so much and congratulations on your film.
S6: Thank you for having me.
S1: That's our show for today. I'm your host , Jade Hindman. Thanks for tuning in to Midday Edition. Be sure to have a great day on purpose , everyone.
