[0:00:00] Shipping, billing, admin, payroll, marketing. You're managing all the things. So why waste time sending important documents the old-fashioned way? Mail and ship when you want, how you want with Stamps.com. Print postage on demand 24-7 and schedule pickups from your office or home. Save up to 90% with automated rate shopping. That's why over 1 million small businesses trust Stamps.com. Go to Stamps.com and use code PODCAST to try Stamps.com risk-free for 60 days.
[0:00:29] Give the gift of the coziest bed with Bolin Branch luxury organic cotton sheets. Only 1% of the world's cotton meets their standards of softness. Treat the people you love to the comfort of Bolin Branch signature sheets, pillows, and more. Feel the magic in every moment. Save 25% sitewide at bolinbranch.com with code BUTTERY. Exclusions apply. See site for details.
[0:00:54] If you've ever sent a cold email, you know how it goes. Figuring out who to email, where their contact info is, and what messaging will land. It takes time, and sometimes dozens of tabs. I've been there. That's why I use Apollo. Apollo is an all-in-one system for data enrichment, sequencing, automated messaging, and intelligence, powered by AI that helps do the work, so you can do everything from finding prospects to closing deals, all in one place.
[0:01:22] Head to Apollo.io and sign up for free. That's Apollo.io. This is Reasons to be Cheerful with Ed Miliband and Geoff Lloyd. Well, hello. Hello, how are you? I'm fine, thanks for your reply. Oh, no, you didn't reply to me. What? What are you talking about? Be careful in that glass house of yours, I've heard. I'm so in this glass house, aren't I? So I sent you an article from The Guardian about ultra-processed food.
[0:01:58] In particular, ultra-processed sandwiches or rather sandwiches, shop-bought sandwiches. And I basically said, look, doesn't this refloat the case for the make your own sandwich thing? But I don't think you replied, did you? I think that I didn't rise to it is how I would frame it.
[0:02:14] You're on very thin ice with this replying stuff. I totally am. But you always have a go at me. Because I would be very happy to do an audit. No, I would lose the audit. But you also do say to me quite a lot, you didn't reply to me. I think it's just basically because you sort of saw the kind of force of my case. And to be honest, I made me think I was an idiot for never having thought before that the case of the make your own sandwich is that you avoid it having all kinds of processed things in it.
[0:02:43] So let me ask you a question. Why couldn't you just start a sandwich shop that sold sandwiches on better bread? Well, made on the spot, I thought you were going to say. Made by professionals. Made on the spot. Who have to adhere to hygiene regulations. A few hours later, I did end up sort of concluding that that was probably the case. Yeah, but isn't it quite common to have a sandwich shop where the sandwiches are freshly made?
[0:03:05] I don't think that's a disruptive idea to the sandwich industry. No, no, no, I'm not saying it is. I was actually being relatively sort of inquisitive and open and just sort of thinking, what's the thing? Anyway, maybe we should just do an episode on sandwiches. Yeah, I think if you were to edit together all the conversations we've had about sandwiches, it may even form a majority of the content we've put out over this past six years. Yeah, I know, it could form a whole box set. But maybe we should just do it and then it'll get it out of my system.
[0:03:33] Would it get it out of your system? I'm not sure it would. It might do. But what would you talk about if you didn't have your go-to of sandwiches? I mean, it'd just be open water swimming then, wouldn't it? Yeah, probably. I've got my cooking. I've got very limited repertoire, haven't I? I think that's what you're saying to me. I made some more dal actually at the weekend. Oh, how did it go? Wasn't as good as the first, the five-hour dal, but it was an hour dal. Can I just say, there are so many different lentils in the world.
[0:03:59] Yeah, it's quite confusing. It's a cornucopia. It's very hard to understand one lentil from another. Or beans. What's the difference between a mung bean and a lentil? I wonder if you're just going too deep into it. I think I am. I wonder if this type of questioning is actually detracting from your enjoyment of making Nadal. Well, there you go. Now we have news. Go on.
[0:04:22] Our news is that we've had a baby. Now, you might have noticed this. Something dropped into your feed on Friday called Reasons Revisited. And if you're thinking, what the hell is that? Why are they spamming me? We are not. This is something we've created for you. And I'm going to tell you a little bit more about it. And this is just like a little trial period. If it goes well, we'll stick with it. And we think it will.
[0:04:43] Every Friday, a little mini episode is going to drop into your feed. It's called Reasons Revisited. And here's how we came to this. We realised that we're sitting on these over 300 episodes of ideas about making our lives and our society better.
[0:05:02] I thought often these ideas are in the news or the underlying issues are in the news. So what we've decided to try is every week we will give you a bite size. And when I say bite size, I mean something that will last approximately as long as it will take for you to be in the bathroom performing your ablutions in the morning.
[0:05:20] 10, 15 minutes long called Reasons Revisited. And what it'll be is a catch up and an explainer on some ideas and solutions to one of the issues that's in the news, that's in the conversation at the moment. It's like a nugget. It is a nugget. A Jeff McNugget. A nugget, a little refresher for you, but not too long because we know that you're busy. So when it appears in people's feeds, they shouldn't think it's a sort of alien life force, but they should think it's not that.
[0:05:49] No, we're not spamming you. This will hopefully be useful for feeding your brains and reminding you of some of the great ideas that are out there that we've looked at over the past few years, which remain as relevant today, sometimes even more so. It's called Reasons Revisited. Look out for it in your feed on a Friday. Great. Well, should we talk about what we're talking about this week?
[0:06:12] This week, we're talking about AI. Over the last few months, we've heard a lot of doom and gloom, I suppose you could say, about AI and the existential threat it poses. Is that justified? We're going to find out more about what the risks are with Mustafa Suleiman, who is a tech entrepreneur who founded DeepMind, a leading AI lab. And then we will ask about whether the hype we hear about AI, both good and bad, is justified with Dr. Vary Aitken.
[0:06:39] And Lauren Goodlad, who is a professor at Rutgers University. What's your reason to be cheerful, Geoff? My reason to be cheerful is the bird feeder that I put in our back garden at the beginning of this year when we did an episode about birds. We talked to the urban birder. I got very excited by it. And for a long time, no birds ever visited. I felt like the birds...
[0:07:03] I'm going to write this down now. The Merlin Bird app allows you to identify the birds from...
[0:07:32] Their sounds or their pictures. I mean, it's like Shazam for birds. First question, does it work? Because I have tried a few of these things and they never do. It does. In my experience, it does. Okay. So what have you identified then? A pigeon. I don't know. It definitely was a pigeon. But no, honestly, it's like really good. I'll download it then. Thank you. What's your reason to be cheerful?
[0:08:00] My reason to be cheerful is that series three of Starstruck is on. The Rose Matafeo show. Are you a fan of Starstruck? I think it's charming. I think Rose Matafeo is a very accomplished stand-up. And from what I understand, I haven't really seen the first episode of the third series, but I think she takes it somewhere slightly different this time around. Is that right? It's basically Rose Matafeo and Nikesh Patel as Tom Kapoor.
[0:08:28] And it's a sort of on-again, off-again romance. I'm not sure they take it that differently, actually. Some of the stuff I've read is, in some ways, the first couple of seasons are a love letter to the rom-com. And then the third series is a bit more, what's wrong with that? Interesting. Maybe we're too early in it to know that. Actually, what I like about it, well, it's light. It's short.
[0:08:54] Yes, shorts, that is an underrated quality. 21, 22 minutes. Perfect. Do you know what it reminded me of? I don't know why we've never discussed this. When I was a kid, I was very obsessed with a sitcom called Just Good Friends, starring Jan Francis and Paul Nicholas. Do you remember? Yes, I do remember that, yeah. About Penny and Vince. Yes. Vince jilted Penny at the altar.
[0:09:19] And then they meet again five years later. Anyway, I was going on and on to Justine about this the other night and I think she wasn't allowed to watch it. Were you just given free reign with that TV when you were a kid? Well, it's slightly strange because I sort of think I might have been, yeah. So if there are any parents listening to this who are worried about screen time... Well, like me, I'm a parent worried about screen time. But look at you, it seems like you had loads of it and you've done fine.
[0:09:41] You've got a podcast. You've got that other thing that you go and do. My parents didn't have any restrictions on how much telly we could watch. I mean, I think they had restrictions on when we had to go to bed, but not really. Yeah, I think the difference was that there was just a restriction on how much TV you could watch because there were so few channels and a lot of the time it was just the test card or pages from CFAX. I do. I really quite like CFAX, actually. Me too. I'm obsessed with it, yeah. Really? Yeah. I think we should tell our younger listeners what CFAX is.
[0:10:10] It's like the internet, but better. And they didn't have it in that many countries, you know, that the Americans didn't have this technology. It was a button on your remote control when TV started having remote control that brought up...
[0:10:23] Pages of text. And it could be anything from TV listings to sports news. There were also magazine features for teenagers or there was teletext art, which was basically portraits of famous people rendered as if they were made out of Lego because the graphics were so poor.
[0:10:41] Do you remember you'd have to wait for the page to come round? So if you wanted to look at something on like page 209, it'd have to cycle through a thousand pages before it got back to the one you wanted to see. I mean, I wonder who invented it. I don't know. We should get them on. I think it was. It did eventually get abolished.
[0:10:57] Yeah, it wasn't banned. It wasn't banned, but it was. It wasn't like this AI conversation where people were... It was sort of got rid of, wasn't it? Yeah, I think my dad was the last person to still be watching it. But up until somewhat recently, he was still watching the football scores on CFAX. I think we're losing any member of our audience under the age of about 45 with this conversation. Oh, Ed, we sound so old. We sound so old.
[0:11:21] We're in a warm bath of nostalgia, but some people just don't want to get in that bath with us. I know. Anyway, Just Good Friends, I did sort of look it up on Wikipedia and had some kind of nostalgia about Just Good Friends. And I think it is a bit the Just Good Friends of 2023, starstruck. Great. They should put that on the posters. Reasons to be cheerful with Ed Miliband and Jeff Lloyd. Now, to begin our conversation, I'm delighted to say that we're joined by Mustafa Suleiman, who is co-founder and CEO of Inflection AI and author of...
[0:11:49] A new book, The Coming Wave, Technology, Power, and the 21st Century's Greatest Dilemma. Mustafa, thanks so much for joining us. Thanks for having me. It's great to be here. Now let's start by asking, what is the coming wave and what is the 21st century's greatest dilemma? The coming wave of AI and synthetic biology are two new general purpose technologies which are bursting onto the scene. I think people will have a...
[0:12:15] I think there are...
[0:12:41] Some characteristics of these two new waves that are different to previous technologies that mean that over the next couple of decades, we have to figure out how to contain and that is limit their consequences and in some sense limit access to those technologies in ways that enable us to prevent the potential threat that they pose to the future of the nation state. What is the greatest dilemma then?
[0:13:08] So the dilemma is that these technologies are essential. We must have them. And they're going to deliver the greatest productivity boost we have seen in the history of our species. I mean, these are the tools that allow us to replicate what has made us so successful as a species. Our intelligence, our ability to synthesize information, use that to make predictions about how the world around us is likely to unfold.
[0:13:35] And then invent new things, create things, solve hard scientific problems, build businesses, et cetera, et cetera. And that is a unique asset of our species. And it has been the engine of progress for millennia. And we're now taking some parts of that process and distilling it into a piece of software. I predict over the next two decades, we're going to see.
[0:14:02] A really incredible boost to our creative process and our general well-being. And I think that there's huge reasons to be incredibly optimistic about those outcomes. But the dilemma is that we can only benefit from those things if we also mitigate the harms, just as we have faced that similar dilemma with the arrival of many other technologies in the past. We've done a great job of mitigating the downsides of aircraft travel or cars or those sorts of things, nuclear power.
[0:14:29] But isn't there an inherent difference in that this goes beyond a tool and AI will have an autonomy and a decision-making capability that's different to any development in the past? Yeah, I think you're right. The difference with this kind of technology is that it has some potential characteristics which we may choose to design into them. These are not inevitable.
[0:14:55] Technodeterministic trajectories. There is no emergence that naturally happens with these models. It's super important to reframe the default expectation that has been set with AI, which is the Terminator framing has led people to think that this is just going to sort of naturally recursively self-improve. There'll be an intelligence explosion. The AI will be able to update its own code. It will be able to create autonomy for itself. It'll be able to set its own goals.
[0:15:23] This is all completely wrong. People may design those capabilities into the models, but there's nothing inherent in the technology, which means that is inevitable. Talk to us about this statement that you signed. One of the people they call the godfather of AI, Geoffrey Hinton.
[0:15:39] Got together some leading minds in the field, people like yourself, Bill Gates, Sam Altman. And it was only 22 words long. I'll read it. It says, mitigating the risk of extinction from AI should be a global priority alongside other societal scale risks, such as pandemics and nuclear war. So why did you feel it was necessary to sign that statement? I have huge respect for Jeff. I mean, he was one of our first advisors in DeepMind in 2011. And I think that it is certainly true.
[0:16:08] Thank you so much for having me.
[0:16:36] In that framing, we do have to take seriously the threats that arise from that. So for example, you know, these models will lower the barrier to entry to most tasks. They will act as a coach, an advisor, a sort of a research assistant, a teacher. And you can think of all the amazing benefits that arise from that. But you can also think of...
[0:17:02] Reducing the barrier to entry to being able to manufacture a biological or chemical weapon. We've seen those capabilities in these new large language models, my own at Inflection AI included. But if our technologies end up in 10 and 20 years in complete open source, then how do we know that some people won't actually use those models to try and do those kinds of things? Those are the sorts of concerns that we all have.
[0:17:26] What does regulation look like, Mustafa, here? Just so we can get our heads around this sort of question. The first thing is that regulators have to be technically competent, which means they have to be able to ask the right questions and probe in sensible technical ways. Take the development of nuclear power or nuclear weapons, for example. I mean, regulators are extremely technically competent.
[0:17:50] So I'm not saying we need that today in AI, but I'm saying that we don't need to reinvent the wheel. There's no need to panic. There are lots of good precedents for regulatory frameworks, which have been incredibly successful. Take aircraft safety, for example. The industry is forced to share insights among competitors if there has been a safety incident.
[0:18:11] We do need international consensus because if we have groups in less regulated areas who are able to experiment without the kinds of oversight that we would want to hold ourselves accountable to in the West and elsewhere, then that does present a risk. Let's talk about your history in this area. So you were co-founder of DeepMind, which eventually became Google DeepMind. I came and visited you after.
[0:18:38] I lost the election in 2015 just to see what it was all about. What did that experience teach you in relation to these questions we're discussing? Well, I co-founded DeepMind in 2010 precisely because I could see the potential impact of this technology. Obviously, I didn't know that everything was going to be as successful as we are, but I could certainly see from that point that even if...
[0:19:03] AI was mildly successful, then it was going to fundamentally shape our values and what it means to be human. And so, yeah, right from the outset, I've been focused on trying to introduce the language and ideas of safety and ethics to the field of artificial intelligence. Is one way of thinking about this, you designed a DeepMind AlphaGo, which turned out to be able to play Go, not just much better than a human, but was...
[0:19:32] Doing moves and thinking in inverted commas in ways that humans just didn't. If you think about medicine, for example, presumably that is the potential that it can recognize patterns around cancer, just as an example that maybe humans have never recognized. Yeah, I mean, look, the quest of artificial intelligence is the discovery or invention of new knowledge. That's what we're trying to do. I mean, you know.
[0:20:00] Civilization is built on us inventing and creating new strategies that help us to live life in a healthier and more productive way. And that's what AIs do. They will ultimately help us in healthcare, in education, in transport, in every possible area, because there are efficiencies to be had absolutely everywhere. I mean, we're in a sort of pocket of inefficiency as a species, and there are huge leaps that we can make.
[0:20:29] We've actually demonstrated it in healthcare multiple times, and we've published many nature papers now in the field on identifying cancers in mammograms, identifying blinding diseases in OCT scans, which is work that I did with the NHS at Moorfields Eye Hospital. So there's lots of examples now of AIs really having a very practical impact. Eight or nine years ago, people were saying self-driving cars are about to sort of take over, and what's going to happen to all the...
[0:20:59] It turned out that the sort of hype about self-driving cars was at least premature. Is there hype around AI, which is premature? I think people always overhype things short term and always underappreciate things long term. You know, I think that's just our bias as a species.
[0:21:20] I think two years ago, AI was probably underhyped because people hadn't wrapped their head around what was going on. Whereas now I think everyone's sort of seen ChatGPT and, you know, they've had a chance to play with these language models. And so, you know, I think there's been a huge shift, but there's always a risk of overhype for sure. Why don't we end with a bit of optimism? If we get this right, what can the benefits be for humanity in the sort of 10, 15, 20 year view? Yeah, I mean, over a 10 year period, for sure.
[0:21:49] Billions of people are going to get near equal access to the most capable teacher, coach, chief of staff, personal assistant, confidant in history. So just as every one of us in the more developed world has broadly equal access to the best laptop and smartphone that humanity has ever invented.
[0:22:14] We're going to see the same trajectory with respect to access to intelligence. And that means that everyone is going to get an incredible lawyer, an incredible doctor, an amazing teacher, all broadly speaking, equally within the decade. And that's why I say I think that's going to be one of the most meritocratic moments in the history of our species. It's going to be incredibly empowering because...
[0:22:38] Suddenly your network, your social privilege, your family, your class, all the things that we have all been concerned about for centuries are going to matter much less. And what is going to matter is your willingness to adopt and use new technologies to get entirely personalized, neatly synthesized access to information really quickly and use those tools to help you do great things in the world and be great inventors and creators.
[0:23:04] Jeff, maybe you will even be able to find a more technologically adept podcast co-host through AI. Oh my God, that would be so amazing. Like an AI ad who knew how to plug his headphones in and mute his microphone when there's a din in the background. That's exactly what AIs are going to do. They're going to augment your weaknesses so that you can spend time being the best ad and then the kind of fumbling headphone ad is going to be a thing of the past.
[0:23:33] Oh, then bring it on. I'm not worried about any of the risks you've talked about if that's going to happen. Christopher Solomon, author of the new book, The Coming Wave. Thank you so much for joining us. Thanks for having me. It's been great to be with you both. 500 orders a month was manageable. 5,000 is madness. Embrace intelligent order fulfillment with ShipStation, the only platform combining order management, warehouse workflows, inventory, returns, and analytics in one place.
[0:24:01] What used to take five separate tools, ShipStation does in one. Automate fulfillment and rate shopping to ship up to 15 times more with up to 90% savings. Go to ShipStation.com and use code AUDIO to try ShipStation free for 60 days. There's one thing that will forever change the way you sleep, and it's the softest 100% organic cotton bedding from Bowlin Branch. You'll fall asleep faster, wrapped in the most luxurious comfort.
[0:24:28] You'll never sleep hot again with airy blankets, cloud-like duvets, and breathable sheets. You'll experience pure comfort on night one and feel your bedding get softer with every wash. Discover the difference with 15% off your first order at bowlinbranch.com with code BUTTERY. Exclusions apply. See site for details. There are two types of business owners. Those who are busy and those who want to be busy. Toast is designed for both with tools to keep you humming and help turn grind into growth.
[0:24:57] That's how you turn busy into business. Toast, built for busy. Next, we welcome back to the podcast, Ethics Fellow from the Alan Turing Institute, Dr. Vary Aitken. Hello. Hi. Thanks for coming back. It's been a few years. I guess what we're talking about today is AI hype, both good and bad. So depending on who you listen to, it's either going to be this huge existential risk or it's going to be completely transformative for society.
[0:25:26] Is that a real binary? Can you place yourself at either ends of that spectrum or is it more nuanced than that? Yeah, I guess I'm far more in the more nuanced kind of middle position. There's definitely a lot of hype and sensationalism around AI at the moment. And a lot of that, I would say, is really serving as a distraction, as a distraction from discussing the realities of what AI currently is, what it's capable of, and also what the real present concrete risks around AI are at the moment.
[0:25:56] Certainly over the last year, there's been lots of announcements, lots of statements about existential risks related to AI. And in most cases, those statements are very well timed to come out exactly at the precise moments where we were having really important discussions around emerging regulatory frameworks, particularly the EU AI Act.
[0:26:12] In most cases, I think that is a deliberate distraction technique to distract discussion away from discussing concrete real risks around AI or how we hold big tech companies and developers accountable for the decisions they're making in designing and developing AI. And to shift the focus to these kind of hypothetical, far-fetched scenarios where the risk is the technology itself rather than the decisions of developers or of organizations, of people. And that's really what we should be focusing on.
[0:26:41] So what's an example of something that's being overlooked in the more kind of grandiose kind of warnings about AI? AI is already all around us. You know, we're all interacting with AI on a daily basis, maybe even on an hourly basis. AI is integrated into the systems that we're interacting with all the time. So at the moment, we're hearing a lot of this kind of excitement, a lot of this hype around generative AI.
[0:27:09] Particularly around large language models. But that's just one form of AI. That's just one category of AI. And AI is already integrated into systems that make decisions about our access to services in the public sector as well as in the private sector. AI is already in the technologies, in our smartphones, online, that we're interacting with all the time. Of course, we need to be anticipating what the risks are further down the line and what kind of advances in AI might enable in the future. But we also really need to be scrutinizing the way that AI is already being used today.
[0:27:37] Can you tell us about some of those decisions it's making and give us some examples? Yeah, I mean, there's AI systems that are used to process information in, say, for example, who gets access to finance or in banking. It's used not necessarily to make those decisions, but to inform decisions. It's used in the public sector to...
[0:27:54] Make decisions about who gets access to services. It's used in areas such as immigration, policing, education, healthcare. And of course, in all of these areas, there are very valid and legitimate things that AI could do in terms of informing decision-making or processing complex information. But the risks are where these systems are used to inform decision-making without accountability, without scrutiny. If they're relied on to make those decisions, or if we don't have insight into what data or potentially what biases are involved in the decision.
[0:28:23] Just to take one of those as an example, you mentioned immigration. So I'm thinking about when my wife applied for her indefinite leave to remain visa. We had to compile a huge amount of information and supporting documents, and then human eyes needed to go through all that to check that she met the criteria.
[0:28:46] That's not the one where you did the wrong Dropbox and then ruined your holidays. I don't want to talk about that. Maybe that wouldn't have happened if we'd had an AI, Jeff, actually. Well, maybe AI could replace me as a husband somewhat successfully. So presumably it has the potential to save the time that you humanise.
[0:29:05] So that is a time-saving use of AI. Are there any examples of AI being able to do things that aren't just time-saving, that are kind of creative and beyond what we would be able to do at our own speed?
[0:29:31] Yeah, and that's where it starts to get really risky because, yeah, a lot of the benefits that AI currently presents are in that kind of realm of efficiency improvements. But to ensure that that efficiency isn't coming at the risks of safety or making inappropriate or inaccurate results, it needs a high degree of scrutiny and oversight. It needs to be checked that there aren't biases within those processes, that it's not perpetuating biases.
[0:29:55] You mentioned creativity, and that's the interesting discussion around the extent to which AI can do something creative, whether it has the capacity to do that.
[0:30:02] And I don't think AI itself can ever be creative. It doesn't have that kind of intellectual capacities to do that, to think for itself. But it can be used creatively. These are tools and they can definitely be used creatively. But it's really important that we always think of them in that way as tools that can be used by people, by humans. And it's always humans that should be held accountable for the outputs that they produce of decisions that are made based on uses of AI.
[0:30:28] How is the UK doing on these issues? I mean, it does strike me listening to you that government tends to be very underpowered on these issues, partly because technology moves so fast and government moves so slowly. Yeah, I often hear the argument that, you know, that can regulation or can government keep up with the pace of innovation? And I think sometimes that argument is used as part of the kind of big tech rhetoric of, you know, that we need to have big tech players to understand what's happening because government can keep up.
[0:30:57] There's no reason why government couldn't keep up. We can regulate AI. We need to regulate AI. And it's not impossible. The UK approach is quite different from the EU approach. So in the EU, there's the EU AI Act. So it's going to be a piece of legislation around regulating AI. In the UK, there's no current plans for a single AI regulator or a single piece of regulation legislation relating to AI. But the approach in the UK is there's a pro-innovation, principles-based approach to regulating AI.
[0:31:25] And that seeks to equip existing regulators to grapple with the challenges of regulating AI across our sectors. You work a lot on thinking about the people who often aren't included in discussions around AI, but could be negatively impacted by it. Talk to us a little bit about this. Yeah, this is something I'm really passionate about. A lot of the challenges, a lot of the problems we face at the moment is that the discussions around AI are really dominated by big tech players. And that's where a lot of these kind of sensational narratives around existential risk are coming from.
[0:31:54] But when we think about the risk associated with AI, it's absolutely crucial that we focus on the voices of impacted communities, people who are really impacted by these technologies, but aren't typically involved in decision making around how those technologies are designed or developed, how they're deployed. One area of work that I work on is around engaging children in decisions around AI. So at the Alan Turing Institute, I lead a program of work around children's rights and AI.
[0:32:19] We're working with primary school children across Scotland to involve them in discussions around AI. And we're involving them to find out, first of all, what they already know about AI, how they feel about AI in their lives, but also how they would like to be involved in the future. It's often suggested that AI is so technical or so complex that...
[0:32:37] Most everyday people can't really understand or can't really get involved in these discussions. That's nonsense. I think that's really quite dangerous nonsense because when people say that, it's really a way of kind of closing down or shutting down debates, shutting down public discussions around AI.
[0:32:50] If anybody comes to the classrooms with these eight-year-old children who are talking about AI, you'll see that they can definitely understand it and they have a lot to say about it. Children are probably the group within our society are going to be most impacted by these technologies in their future lives. But there are many, many groups who are underrepresented in design processes and development processes. And it's really important that we bring them into these conversations. Just to end with, we have something on the podcast called the Jeffocracy. Jeff is the...
[0:33:17] If you became the minister for AI, maybe, what would be your first act? Well, yes, I have lots of ideas for this one. But yeah, the main one would be that I would want to create like a citizens panel, citizens assembly on AI.
[0:33:34] At the moment, there's a big focus around the government are planning the Global AI Safety Summit. But it really is, from what we've seen so far, is really centering the voices of industry, of big tech, and really focusing on the safety of future advances in AI. And what I would want to see is something which is really focused on the real current risks of AI, looking at ethics around AI.
[0:33:57] And understand what matters to people when we think about the future of AI. How can we harness the value and the benefits of AI and show that they're equitably distributed across society? And also understand what wider public concerns are around AI and how we can address those. Did she get the job, Geoff? Well, I will feed that into the AI and see whether it thinks she should get the job or not. Variedkin, it's been a pleasure to talk to you. Thank you so much for joining us. Thank you.
[0:34:23] So to complete our conversation, I'm delighted to say that we're joined now by Lauren Goodland, who is Professor of English and Comparative Literature and Chair of the Critical AI Initiative at Rutgers University in the United States. Lauren, thank you so much for joining us. Thank you for having me. Maybe you could just start by telling us a little bit about yourself and how an English professor finds herself chairing a Critical AI Initiative at Rutgers.
[0:34:49] So I'm going to give you the very quick version of this because it could go on forever. But I'm essentially a historian. I'm a Victorianist. And so I started working on the history of statistics and realized that many of the basic tools that are fundamental to how AI is designed today were Victorian era innovations in statistical modeling.
[0:35:18] And they were done by famous eugenicists like Francis Galton and other people that were interested in what are now considered to be pseudoscientific ways of thinking about biometrics and modeling people. And I realized that people who work in the technology sector in fields like AI ethics, they too had been interested.
[0:35:46] So I began collaborating with technologists. I worked very closely with several AI researchers. And here I am. So you're saying that AI is built on a bad foundation of extrapolating poorly about human beings from data? I don't want to call it a bad foundation, because I actually think that it is sometimes exactly the foundation that we want for some things.
[0:36:13] But it all depends upon the quality of the data and its suitability to model what it is that you're looking to model. And it is a statistical model. And in fact, that's really what we're calling AI today should really be called data-driven predictive analytics because that's what it is. The world is full of data.
[0:36:40] Or at least the parts of the world that use the Internet are. And that data is taken from Internet, from Internet of Things, from devices, from sensors and cameras that are all over the place, largely without our consent and often in violation of copyright. So that is one ethical dilemma. And Lauren, tell us what your critical AI initiative at Rutgers does.
[0:37:08] Well, we do a lot of different things. We're highly interdisciplinary. We're also very international. The reason that we're working with data scientists at Pretoria is that one of many problems with the language models about which there is so much enthusiasm today is that they are preponderantly trained on data.
[0:37:33] I mean, bear in mind that 30% roughly of people in the world have never used the internet. And what this means is that in spite of the fact that the amount of data on the internet seems enormous, it enormously over-represents the people on the internet.
[0:37:59] Who are preponderantly white, preponderantly male, preponderantly North American. And that is one of the reasons why bots mimic and pick up so much, to use the technical expression, garbage in. And that is why they need so much instruction, that's a term used by the industry, by people, usually very low-paid people.
[0:38:26] Working from the global south, places like the Philippines or Kenya, to at industrial scale, label and annotate the conspiracy theories, the pornography, the disturbing stories that would embarrass companies that build these tools. We're producing a monoculture. And when you put that together with the first thing I said about how statistical modeling is really the genie, the secret sauce.
[0:38:55] Behind being able to predict what the next word or sentence might be if I put in a prompt. And you realize that that works by picking what is the most probable. What you realize is the degree to which this is normativizing language.
[0:39:19] To the point actually even of potential model collapse. It lacks the diversity to simulate how people talk about the world. It's just a model, and it's a probabilistic model mimicking how a certain swath of people who are hugely overrepresented are going to talk about the world in answers that are the most probable of all possible answers.
[0:39:47] Lauren, listening to you, it's quite a pessimistic account of the impact of AI. But you did say earlier that you thought there might be some benefits. I mean, do you see a kind of well-regulated advance in AI as potentially offering big benefits like the medical benefits, for example? Absolutely. I am not what is conventionally called a Luddite, although let me put in a plug for the Luddites. They were not a good thing.
[0:40:15] They were actually fighting for economic justice, not against technology. So in that sense, I guess I am a Luddite. But the use of predictive analytics to help problems for which it is suited could be amazing, right? And certainly it can help us to build better drugs, say, through technologies like protein folding, better weather prediction.
[0:40:44] Hopefully climate models that get us to do the right thing, possibly make energy more efficient. The issue is that the world's most powerful and lucrative tech companies who are dominating the research in this area, they are not doing those things. These companies did a lot of more blue sky stuff, but mostly what they're all in on now is language models.
[0:41:11] I find it very unfortunate that these powerful tech companies that have in the past tried so hard to portray themselves as sort of public spirited companies that really want to do right by the world, that they are so completely uncritical about the copyright infringement, the carbon footprint.
[0:41:36] I just read a story yesterday. Microsoft does the training for OpenAI. So Microsoft's use of water to cool data centers for training AI has gone up by 34% in one year. And Google's by 20%. And there's not many of them. So of the many issues that I've discussed, one of the...
[0:42:02] One that needs the most regulation is that these companies are just too large and too powerful. Can we talk a bit about creativity and AI? Because I've been thinking about this quite a lot in the light of the Hollywood writer's strike. Because one of the issues...
[0:42:19] What about if AI could write a show better than Succession?
[0:42:45] on its own. Because you mentioned the Luddites before. And I was thinking, my problem isn't with the machine loom. That's great. It feels like an efficiency that stops humans having to do something mundane. It's about that those people then get to continue living a life of purpose and financial security. But it feels different with creativity, because it doesn't feel like you're freeing a writer from a mundane task so they can live their best life.
[0:43:15] That's a great philosophical question for a creative writer to ponder. I'm sure Ishiguro is busy at work on this very novel or something like it. We're talking about the most gifted writers in the world. These systems will never be able to do that. Now, could there be some kind of wholly different technology?
[0:43:42] I think it's a philosophical question, but you're absolutely right. Nobody asked for machines that would write things. And right now,
[0:44:11] You put your finger on where the stakes are. A studio knows that an AI system is going to produce a garbage screenplay. But if they can hire a writer to revise rather than to do a first draft and save money, why not? Let's end Lauren. There's clearly a lot of dangers and risks here that you perceive. Is there something cheerful you'd like to leave us with about the work you do?
[0:44:41] People support us. People want to be empowered. There's a lot of people that do not buy the narrative, which is that this is going to do great things at some point, but it could also do terrible things. And we've got to regulate it because the stakes are huge. But what's actually going on right now, don't worry about that. We're not even there yet. And if this begins to seem confusing and disabling,
[0:45:10] It is because it's a sort of an incoherent mess of warnings and hypings and hypings that seem like warnings and warnings that hype. And all of it makes people feel dumb and disempowered. Anybody can understand how AI works. It is not, to use the cliche, rocket science. It is statistics.
[0:45:40] And right now, those models that are being sold as the work of geniuses and all of those people that are labeling and picking up the pieces to sort of mitigate the problems, that's the system. So I think that it's going to take time, as with any new technology. I can't tell you right off the top of my head how long it took to make trains safer.
[0:46:07] Airlines, you know, think about how heavily regulated they are. Think about what it would be like if Boeing or Airbus said, we've got this great plane. It's new. It's better than ever. But we're not going to tell you how we built it. We're not going to let other engineers come in. I think that's good news because it means that we're going to need to work together, especially younger people.
[0:46:34] To empower ourselves as citizens and demand the kind of regulation that says, this is not the AI that we need. We need AI that serves the public interest. Well, look, Lauren Goodlad, it's so interesting to talk to you and to hear your perspective looking at AI from a completely different point of view. Thank you so much for joining us. Thank you. Well, are you looking forward to having your AI podcast co-host to replace me? I think augment you. You sure? Yeah.
[0:47:03] I'm seeing some kind of Ed AI hybrid cyborg thing. I think you'd prefer that. I think that AI chatbot thingamajigs would be better at responding to your text messages. We've had a few conversations about AI at this point. And for me, this is perhaps the best one. Yes. Thinking about it in a slightly different way. And I say this as somebody who is... You're a techno enthusiast, I would say, aren't you? Curious. Yeah, yeah, yeah.
[0:47:28] Have you listened to any AI music? No. So on TikTok, I came across something called the AI Beatles. And it's basically somebody has taught AI to sound like the Beatles singing and has generated a song. And I listened to this at the weekend. And it is about as good as a quite good tribute act. So it's not identical, but it has trained it to sound like their voices. And there's more and more of that.
[0:47:57] It's a big conversation in the music industry. And that sort of gets you onto copyright and scraping. And then I can back and forth a little bit on that because isn't that what we all do when we're learning something? So I think quite a lot about some of the issues around this. And I think up until this conversation, I'd kind of had a, some of it's not great, but do genies ever go back in bottles?
[0:48:24] But then hearing explained in talking about the way we regulate, say, aviation. Yes. Or the other one I was thinking about actually was cloning. Yes. Because, you know, I worry about what somebody with a supercomputer powered laptop in 20 years time that must have always talked about what they could do sort of if they went rogue. But that doesn't really happen.
[0:48:49] That's a really good point.
[0:49:03] It could be regulated and contained, to use Mustafa's word. And then I was just so interested in all the stuff about it being a language model and whose language is that scraping, how it's representative of the type of people who have historically been more typical to use the internet over the last 20 years. And then I was quite struck by...
[0:49:27] Something that Lauren said there about how language is only part of how we process. So if you think about what the inputs are into a human brain, it's not just that information that could be scraped from the internet, either in terms of just knowledge or understanding how we work. We work in these incredibly complex ways and no inputs for any AI system could come close to that, I don't think. I mean, you've got really interesting things to say.
[0:49:54] One thing that I take out of it is this point that it's easy to just focus on, oh, what is this existential future going to be? You know, is this going to be a risk, et cetera? And that's all okay. But AI is being used at the moment.
[0:50:12] And is it properly regulated? Is it properly monitored? You know, I think we can spend a lot of time thinking about what's going to happen in 10 years time without thinking about what's going to happen now. Yes. And how are these first movers making sure they do a land grab in such a way that sets them up for the coming 50 or 100 years? Well, also, the way AI is used at the moment, that is surely going to define how we do it in the future. So in a way, we could spend a number of years discussing how we want to regulate this.
[0:50:39] I sort of feel it's a fascinating conversation. And I feel above all, you can't sort of stick your head in the sand and pretend this isn't happening. Because enough smart people think it's going to be quite transformative. And I also do think about medicine and think about how it could be positively transformational. That's really exciting. The finding of patterns that, you know, humans can't identify.
[0:51:07] But then, you know, making sure that that foundation that it's built on doesn't contain biases that miss stuff or miss stuff for certain groups of people. Completely. Send us your ideas or suggest a guest for a future episode. Email reasons at cheerfulpodcast.com. Find us on Facebook or tweet at cheerfulpodcast. Whoa-ho-ho, we're in the outro, ho-ho. Do you want to hear something disgusting? No. I won't tell you then.
[0:51:36] Go on then. Had a bath the other night, got out of the bath, went into the bedroom, looked down, my entire big toenail had fallen off. It was just there on the floor. Why is that? From when I had the bike accident. Oh dear. I think that was one of the points of impact and it's been slowly working its way off. That's your material for the outro, is it? Well, I was going to ask you advice. I'm struggling to think of a 10th anniversary gift for my wife for our wedding anniversary this week and I was thinking, what if I put it in a locket? Honestly, that's a terrible idea.
[0:52:08] Okay, I won't do it then. Do you want it? No. What am I going to do with it? Throw it away, honestly. It feels like a terrible waste. No, honestly. Put it in a time capsule and then future generations will uncover this time capsule and will be really disappointed. They might be able to clone me from it. Yeah, I mean, after listening to this episode and listening to Mustafa, for example, maybe that is anything that's possible. Okay, well then what I'm hearing is you think I should definitely keep it. Thanks, Ed. Yep, no problem.
[0:52:36] A reminder to look out for Reasons Revisited in your feed on Friday. I'd like to thank our guests, Mustafa Suleiman, Vary Aitken, and Lauren Goodlad. Emma Corsham is our audio producer. Rachel Barmer is our content producer, supported by Joe Kenyon at Goldfish. Gail Loftouse is our announcer. James Deacon made the idents. Ed Seed composed the music. And our artwork was designed by Henry Cull. He's been Ed Miliband. He's been Geoff Lloyd. And these have been Reasons To Be Cheerful.
[0:53:05] So when peak season hits, you're ready to go.
[0:53:34] If you've ever sent a cold email, you know how it goes. Figuring out who to email, where their contact info is, and what messaging will land. It takes time, and sometimes dozens of tabs. I've been there. That's why I use Apollo. Apollo is an all-in-one system for data enrichment, sequencing, automated messaging, and intelligence, powered by AI that helps do the work, so you can do everything from finding prospects to closing deals, all in one place.
[0:54:02] Head to Apollo.io and sign up for free. That's Apollo.io. You've tucked everyone in. Now it's your turn. Bolin Branch makes luxury organic cotton sheets. Only 1% of the world's cotton meets their standards in softness. Pull a cozy blanket over buttery soft signature sheets and enjoy a little comfort of your own. Save 25% site-wide at BolinBranch.com with code BUTTERY. Magic in every moment.
[0:54:32] Exclusions apply. See site for details.