The incredible stakes of Elon Musk's trial against OpenAI

0:04Hello and welcome to the Times Tech Podcast where every week we unpack how technology is reshaping business culture and everyday life.

0:12Danny Fortson: I am Danny Fortson out here in the valley covering one of the most dramatic tech courtroom trials, Elon Musk versus OpenAI or Sam Altman. And I'm Katie Prescott here in London, where there are also wild storms going on. Some pretty big political ones and weather ones, I should say. But of course, the tech drama is over in California, where so far we've had two newly revealed billionaires, secret informant allegations and, you know, just some plans to build cities on Mars too. So this week is a reminder that while we are busy looking ahead to the future, where AI agents may replace humans and data centers could be built in space and all of that sort of stuff.

0:58Actually, the messy corporate origins of AI continue to haunt us. Yeah, and the origin of OpenAI is exactly what Elon Musk is questioning in this trial. It's now entered its third and final week in terms of testimony. So a verdict could come as soon as next week.

1:17Danny Fortson: And it's really laid bare the conflicts and rivalries lurking behind open ai's rise from little non-profit to 852 billion dollar company just a little startup as we like to refer to it

1:36pretty startup open ai but i just wanted before we get into the trial and there is so much talk about there i wanted to tell you about this lunch i had last week oh yeah it just made me chuckle because I was just in a bog standard restaurant in King's Cross. And on the table next to me was this collection of, I think, Americans and French people who were no joke talking about building data centers in space. Well, you know, these are the times we live in. It's not a conversation you hear very often in London. And obviously it was because King's Cross has become the tech center in London. You've got obviously Google DeepMind there, all of the british startups like wave driverless cars but also open ai and anthropic as well yeah

2:23Danny Fortson: but it's just not those conversations in the tech hubs i was talking to our producer yesterday marnie and she was in a sauna as one does sauna sauna ing if that is the verb at the moment in the uk and there's a guy talking to his friend about his open claw agent and marnie overheard that this ai agent this open claw could sense from his kind of his human's social media activity that he was missing his former person his for his ex and so he claimed that uh the agent took the liberty of messaging his ex on his behalf i can see this is worry i mean i do worry that norman has access to 20 years of my gmail your your agent yes my ai agent i should say sorry my my open floor norman sits on top of that stuff and you know has potential to cause havoc we obviously can't verify if that happened but a post has gone viral on reddit after someone claimed that their ai agent tool sent messages to their ex without their knowledge so maybe they just read it maybe this is the person and then they post about it who knows so you know between your lunch and people kind of like having their their bots do their bit well not even do their bidding kind of go behind their back i mean these are these are the times we're living in 2026 it's weird weird world and if you listen to that special episode on ai agents and when we were talking about the genesis of Norman recently, you'll know that Norman lives on a Raspberry Pi computer.

4:05And I'm very excited that we've got Eben Upton, the founder and CEO of Raspberry Pi on the program today to talk about how he started the business and what it looks like in this new era where Raspberry Pies are the place where open floor agents are living. But let's start with the trial. Just remind me how we got here like what what is this mad crazy trial with all of these billionaires all about so i can sum it up in a line this is a dispute effectively about money and power and the question at the kind of core of it is is this a case of the most sour of grapes or did sam altman actually steal a charity from elon musk open ai was founded in 2015 by sam

4:55Danny Fortson: maltman ilia suskover greg brockman elon musk and a couple others as this non-profit for the benefit of humanity we're going to build a digital god and we're going to make sure it's good for all of us fast forward to two years later they all kind of come to the realization at the same time oh the bigger computer you have the more powerful the ai becomes what that meant was they needed mega mega mega supercomputers which cost billions and billions of dollars something a charity would have a really hard time raising through donations so they start talking about like let's reconstitute this into a new thing and that's when everything broke down elon musk wanted control he wanted equity control of a new for-profit kind of entity that would be like a sidecar to the non-profit They said, no, we don't want an AI dictatorship, blah, blah, blah, blah.

5:50Danny Fortson: It all got very embittered. Elon Musk left. He had been the sole benefactor up to that point. He had given over$20 million in cash and$18 million in rent payments. And so he left and said, you have 0 % chance of succeeding. And then, of course, Sam Altman did the unpardonable. He succeeded without Elon Musk. And he raised$175 billion, more than any company has ever raised ever, to create what is today OpenAI, ChatGP, developer, et cetera, et cetera. And so Musk is basically saying, hey, I donated all this money to a charity. I left. You turned it into a for-profit, created this giant company that's now one of the most valuable on the planet.

6:41Danny Fortson: That's not fair. you know basically he wants to destroy open ai send it back to its status as a non-profit they say this is ridiculous so the the trial has really been about that key period 2017 2018 when they realized we need lots we need many billions we need to reorganize this organization and that's when everything fell apart and there has been just i mean the most extraordinary group of people coming through through the courtroom and and giving evidence so i think last week we spoke about elon musk yeah but since then well who have we saty nadella the boss of microsoft so major he was there on monday yeah and then sam altman himself so who have you seen and and what have they been saying so i i think one of the things that has struck me sitting in this court day in day out is it really kind of drove home the stakes of all of this because it felt like every person who took the stand and it's not every person but it's a it's a stunning amount everybody who's taking the stand is a billionaire it's kind of wild so it's like greg brockman he's one of the co-founders they're like your stake is worth about 20 billion ilia satskover who's like the star scientist who they prized away from Google.

8:03Danny Fortson: He is worth$7 billion. Satya Nadella is worth over a billion. Brett Taylor, the chairman of the board, he's worth a couple billion. Mira Murati, also a billionaire. So she was the former chief technology officer, is that right? So when you step back, you're like, oh my goodness. Like all of these people, and a lot of them are very young and have kind of come from nowhere to be, you know, many billionaires over, you're kind of like, oh, wow, these are the stakes. This is why there is a trial. Because whomever is in control of this technology, whomever is on that ground floor, stands to make an extraordinary amount of money and have extraordinary power.

8:48And I guess a lot of the trial hinges on what Elon Musk knew. Yes. what he knew and what he didn't know what's come out about that so basically it's again this is all

9:00Danny Fortson: about 2017 2018 they start talking like okay we need to do something else other than a charity if we're going to raise the tens of billions we need and they start talking and there are there are records of this there are text messages there are emails or even term sheets bandied about around okay we're going to create a new for-profit here's how it would look we need to raise at least 10 billion dollars blah blah blah blah blah um and in that elon musk was like look i've been the funder of this non-profit i feel like i should one have majority control of the equity and be the ceo and he's like don't worry though if you're worried about one person being an ai dictator i promise to like dilute my stake quickly over time but there was no guarantee like there's nothing in writing that said he had to do that.

9:54Danny Fortson: And on the stand yesterday, Sam Altman was like, Sam Altman ran White Combinator. They funded thousands of startups. He's like, my experience is when there's a startup, and especially when that startup's going well, people aren't going to be like, you know what? I'm good. I'm going to take my hand off the wheel and hand control over to somebody else. That's just not a thing, and especially not in the case of Elon Musk, who apparently had said to people at the time, I'm not doing any companies anymore where I don't control them. But at the end of the day, they just couldn't agree. They didn't want Musk to have total control.

10:32Danny Fortson: And he sent this very now infamous email where he's like, basically, without me, without my support, I give you a 0 % chance of success. Not 1%, 0%. He actually said not 1%, 0 % chance of success. And he basically left OpenAI for dead. And that's when OpenAI turned to Reid Hoffman of LinkedIn and Microsoft. And all of these other people. And they made the point like, you know, after he left, they started raising money. They raised something like 90 million from all these different sources, from all these different donors when they're still a nonprofit, which again is more than Elon Musk gave. And they're like, are you being sued by any of those people?

11:16Danny Fortson: They're like, no. It was very interesting. And I'll say one other point about this. Ilya Sutskova, he is like the key man in terms of the science. He was at Google, and he was like, they loved me at Google. They didn't want me to leave because when OpenAI was starting, they're like, we need Ilya Sutskova. He is the key because he is this brilliant person. And they're like, right before he resigned to go to OpenAI, at this fledgling lab that was where he was being heavily recruited by elon musk by sam altman by greg brockman they're like they didn't want me to leave they offered me an extraordinary amount of money six million dollars a year which back in 2015 just for like a random scientist at google deep mind is like quite a lot of money he said no and he he recounted that email from elon musk the zero percent chance email and the judge she hasn't asked many direct questions to anybody testifying but she asked him that she's like so was that true like did you have a zero percent chance of success was that how it felt and he said he he compared the ai that they had developed at the time back in 2018 to an ant he's like today and his delivery was amazing he's like today it is a cat then it was an ant and i was like what what like that's that's why google needed him exactly exactly but then i i did i did do this quick look up on claude actually anthropic i was like compare the neurological power of a cat versus an ant and it was actually quite informative so a cat has 250 million neurons and about 300 million in the cerebral cortex blah blah blah and ant has 250 000 so there is a thousand fold difference in just the neurological horsepower between he was he was making a comic a quite profound point i am almost certain i don't know him but i'm almost certain that he was like i'm going to choose these two things because there's that thousand fold difference in the neurological power but anyway i thought it was quite funny the other thing that people are watching from this trial of course is what it tells us about the character of sam altman and it comes at a particularly interesting point because it follows on from that very very lengthy new yorker piece yeah yeah where two journalists spent a year profiling sam altman doing lots of interviews about with him but also with people around him and criticized him really for lying, something which he's obviously denied.

14:01Has anything come out around that in the trial? And is that something that Elon Musk has tried to play on at all?

14:07Danny Fortson: Yeah, his lawyer, Stephen Muller, on cross-examination yesterday, it was really interesting. He's like, should anyone believe you? Do you tell the truth? Do you always tell the truth? And like Sam Altman, he's kind of like, I believe I'm a truthful person. He's like, should we believe you? Should the jurors believe you? Do you tell the truth? And then he went through like a laundry list. He's like, well, Tasha McCauley, she was on the board. She accused you of lying. Did she not? Oh, and he was kind of like stumbling. And he was like, what about Helen Toner? She accused you of lying. Ilya Suskever, he accused you of lying.

14:46Danny Fortson: Elon Musk. And he just like went down this list as like a slow motion character assassination, which was really, I think, effective in terms of what he was trying to do. But it gets back to that core question of like, is this sour grapes or like him being deceptive doesn't mean necessarily he stole a charity in a way that is provable in a court of law from Elon Musk. you know so but it was like it was quite uncomfortable to watch sam altman up in the witness box being like like the first 20 minutes of questioning was just like going through a list of people who he's worked with closely who all accused him of some form of deception sam altman and open ai of course strongly dispute these claims and sam altman has also disputed the allegations in that article in the new yorker so the biggest question though is when will we get the verdict?

15:43Maybe next week.

15:44Danny Fortson: Today, as we speak, is the last day of evidence. And then they're going to start deliberations. I believe either tomorrow, Thursday, we're courting Wednesday or Monday. And again, very important. The jury is advisory. They're going to give their verdict, but it is up to the judge to decide. So we'll see. But I think, you know, Now, Musk wants them to basically unwind the for-profit, go back to a for-profit entity, give up$150 billion and remove Sam Altman and Brockman from the board. In other words, he wants to destroy this company. Keep in mind, three years ago, he founded a direct competitor in XAI.

16:26It's just the most astonishing case. And it's like just one of the, we've said this before, but a thousand times, but it is extraordinary to see under the bonnet as well of all of these businesses and to see these emails and to see these private diaries and hear the conversations. Okay, well, should we move on from what the billionaires are doing in court in San Francisco? As fascinating as it is, and I guess you're going to be on tenterhooks now, right, for that verdict? Yes, indeed. Yes, indeed. It'll be a sort of Danny get to court now moment. Yeah, they did say, we don't know when the verdict will come.

17:02But typically, you'll have half an hour, you have 30 minutes to basically and they're going to send out like a email klaxon, all caps, basically verdict in get here as quickly as possible. Okay, so don't leave your phone now. Exactly. For a week or so. Let's move on to our guest today, who is Eben Upton, the chief executive and founder of Raspberry Pi and complete move away from what you've been looking at in court. Eben is a very, very Cambridge British tech entrepreneur. And his business, Raspberry Pi, makes these very cheap, sometimes as cheap as£10, very, very small, sort of almost credit card sized computers.

17:46And it's a company that he started back in 2008 when he was director of studies at a Cambridge college, desperately trying to get more people to apply to do computer science. You don't have to try very hard now. It's like one of the most popular courses. But anyway, different era. And it's having a bit of a moment right now because people like me are using Raspberry Pis to house their agents because they're cheap and they don't use huge amounts of energy. Raspberry Pi actually listed on the London stock market a couple of years ago, rare for a tech business, and it's now worth about 1.4 billion pounds.

18:24Welcome, Eben. Thanks very much for joining us. Well, thank you for having me. um eban so i'm way out here on the west coast i've known the name raspberry pi and kind of what vaguely what it is but give me the origin story i'm a child of the 1980s and i learned to program on a bbc microcomputer which was this this incredible idea actually that that your state broadcaster would have a microcomputer very 1980s thing and there was one in the corner of my classroom. And later on, there was one in the corner of my bedroom. And there was never a day when I decided I was going to become an engineer. But you kind of just got lured into it by the kind of ambient programmability of your environment.

19:08And I arrived in Cambridge 30 years ago this year to study physics and computer science. And everybody I was surrounded with on the computer science course had had exactly the same introduction to computing as me, right? Because we historically haven't been particularly good at formally teaching computer science to children. I think we're better, interestingly, we're much better at it now than we ever were before. We're much better at it now over the last 10 years than we were back in the 1980s. So the universities and industry had this amazing supply of people who'd kind of trained themselves to be computer programmers.

19:40I've been programming a computer since I was eight years old. So I've been programming a computer for a decade by the time I arrived at university. And then what happened during the 1990s was those machines kind of disappeared, mostly disappeared. And the PC and the Mac are kind of the only surviving examples of real general purpose computers in people's lives. And they're not particularly programmable. You can choose to program them, but they don't have that architecture where you turn them on and they go beep and they give you a programming prompt. So you've kind of gone, you went into this world where young people were much less likely to be exposed to programmable computers.

20:16That led to a collapse in the number of applicants to study computing at university. It was also leading to a decline in the number of slightly older young people available to go into industry. So if I were to go out onto the engineering floor here at Raspberry Pi, you'd see there's kind of a gap. There aren't a lot of 30-something-year-olds. There are quite a lot of people. Once you get into the early 40s, you start to encounter the people who had my kind of experience in computing, but there was certainly a window of time where people weren't becoming engineers. And really, Raspberry Pi, we started the Raspberry Pi Foundation in 2008.

20:52And the goal was to provide people with a piece of programmable hardware, which was affordable, so affordable, programmable, robust, and fun. And we ended up making this little Linux computer. We launched it in 2012. So then we go on this kind of strange journey through actually into education in a very indirect way. So into education via hobbyists. We sell a bunch of units in 2012, mostly to adults who already know about computing. But then gradually through diffusing it into that community, it then starts to leak out and it leaks out into education. And we've had some, you know, we've had, I think, some good impact in education, but then also has leaked out into industry and then has really led to the commercial side of Raspberry Pi becoming more of an industrial electronics company.

21:46And for people who don't know what they look like, they range from a sort of small credit card circuit board style Raspberry Pi. Is that right? For sort of 40, 50 quid? Yes, these little, I think the cheapest one's about 10 quid. But yeah, these little sort of things that are about the size of a credit card. And I mean, obviously, because the BBC Micro is a, All of these 1980s computers that we were inspired by have that same form factor. It's a computer inside a keyboard. And so once we, when we were in 2012, when we were, you know, two guys in the shed, all we could make was a little PCB. And so you would plug a mouse and a keyboard and a display into it.

22:27And then as the organization became a bit more sophisticated, we grew the ability to do, you know, industrial design injection molding. And we did, you know, from 2020 onwards, we've been offering something ever so slightly before Apple and about a week before Apple launched the first ARM-based Mac products. We launched this product that we called Raspberry Pi 400, the first generation of a Raspberry Pi kind of built into a keyboard. How many are out in the wild now, roughly? It's roughly 75 million Raspberry Pis. 75 million. numbers and it ends up being after the pc and the mac um it ends up being the third most popular general purpose computing uh architecture in the world but you're having this moment this this agent moment can you talk about like how you because as we talked about i've been in this courtroom in california the last two and a half weeks and it's all about kind of the future of humanity what ai is doing what it's gonna do and all these like really big weighty kind of subjects.

23:34How do you see this moment from where you sit? I think we've had a surprise, haven't we? We're about 15-ish years out of the last AI winter. So AI has got, goes through these, went through these winters. Yeah. Where people became very demoralized. You know, people start in the 1950s thinking it's going to be fundamentally fairly trivial to build computers that can at least do tasks like look at a scene and describe the objects. And people go through these sort of periods of great enthusiasm about different models of how AI is going to work. And we're about 15 years into this current period, where it's become apparent that if you bring together some fairly old ideas about trying to build neural-like structures, structures which are inspired by the structures of the brain, with very large amounts of computing power and critically huge amounts of training data, that you can build systems that do really complicated things, slightly worryingly systems that do very complicated things whose creators don't fully understand, don't understand how they do those things much better than we really understand how our brains.

24:43So you have this kind of partial understanding of how these systems work. So we're kind of in this interesting era, probably went through 10 years where this was mostly a story for kind of general purpose users of AI, was mostly a story about image processing. And then suddenly in 2022, you emerge into this world where, you know, the LLM line of thinking, the large language model line of thinking suddenly bears fruit with tools which feel like they do things that general users can get value from. So we've kind of come through this kind of bot era. And now we're into this agentic era, this idea that you can then take some of those capabilities that have grown up, give them access to your, give them an ability to iterate, give them access to your, to other things, you know, to your, to your local environment, give them access to your, to your data, tell yourself a story of some about security.

25:42I think people are trying to find out, people are really at the moment trying to figure out what story they want to tell themselves about how comfortable they are with how much of your uh you know how much of your your data you let them see um and you know what what the security what the opportunities are what the functional opportunities are but also what the security threats are from this and when did you first see the impact on sales at raspberry pi because a lot of people are using raspberry pies as a cheaper way of running agents not only for buying the product but also because they use far less energy than, say, a Mac, which is the other, a Mac Mini, which is the other thing that people are using to run OpenClaw.

26:21Yeah, I think, honestly, I'm not sure we have seen an impact on, I'm not sure we have seen an impact on sales. You know, there are clearly millions of instances of OpenClaw running on Raspberry Pi hardware. And I'm sure that we've sold incrementally tens of thousands, at least tens of thousands of Raspberry Pis to people. So people, to kind of recap, all of these tools, you can run them on your PC. They're just software. But one of the stories people tell themselves about security, one of the arguments that people make is that these are sufficiently powerful autonomous systems that you benefit from putting them on some dedicated piece of hardware.

26:56And obviously, the Mac Mini and the Raspberry Pi are the two classic pieces of hardware that people have chosen to run claw software on. Now, I'm sure we've sold tens of thousands of new units. But I think what you have to remember is almost everybody who's inclined to run OpenClaw already owns a Raspberry Pi. And so I think many, many, many of those units are running on Raspberry Pi. It's not an especially demanding piece of software. So it's not like you need to go absolutely to the bleeding edge of our product line in order to be able to run one of the claw platforms.

27:25Danny Fortson: You talk about coding going back to when you're eight years old. How do you think about this moment in terms of like just, you kind of went over that trajectory, but, and we're in this moment where we're moving from these kind of like really amazing answer engines to these like action engines these things that can go out into the world and do stuff um do you feel like you're living in the sci-fi dream of of young evan or you know good bad indifferent are we sure we'd be completely freaking out like some people are i i don't understand why people are freaking out honestly i mean maybe i'm just slow right but i i don't understand and there is a say more about this because this is a big because like i feel like out here it's marketing by fear it's like white collar bloodbath we're all going to be turned into paperclips we're going to be enslaved by the mission this intelligence that is going to be so much all of that stuff there's this we play this game of what's the punchline and you know maybe these tools aren't any good for anything apart from writing apart from writing stories about how amazing they are that's the main thing people use ai tools for is to try to move stocks around Matt Levine has one of them, Bloomberg, has this wonderful term business negging.

28:43He characterizes Sam Altman in particular, his style of marketing as being business negging. If you say to somebody, my company makes amazing technology, people think, yeah, well, everyone's, you know, I'm not going to invest. All the businessmen say that. But if you say my technology, I'm really worried about it. It might kill everybody on earth. People think, wow, OK, that must be incredible technology. I'll definitely invest in that. So obviously, as you say, there's kind of a marketing through fear element here, which is either conscious or unconscious. But I think for a lot of the people who are advancing the user, I think they're very sincerely held views.

29:17And I think people are caught up in this kind of contagion. So people should still learn to code, in your opinion? People should learn to – well, people should learn to program computers. I've always been nervous. I've always had this personal nervousness about the use of the word code. And I haven't really quite understood or been able to articulate properly why I was nervous about using the word code for this thing that we hope that young people will get excited about. And I think that it boils down to the idea that code privileges the typing. People say, wow, these LLM platforms, these agentic platforms, they're great at coding.

29:59They are. They're okay at coding. But yeah, it is incredible that they can code. But what people forget is that that's actually a very small amount of the work. When people say it's going to put all the computer programmers out, well, they say it'll put all the coders out of work. Coding, sitting down and typing in computer programs, is a very small part of the day-to-day life of a software engineer, of a computer programmer. And so this kind of puts some limit. If you made the coding bit, the waggling your fingers and typing and hitting keys bit, infinitely cheap. if you made, which the tools don't do, but if suppose they did do that, it still wouldn't get rid of the actual job of a computer programmer, right?

30:42And generally sort of you can argue that the tools might become sophisticated enough to eat all of the other bits of being a computer programmer, but those at that point are basically AGI. That replacing all of the computer programmer basically requires you to build an AGI or a human or superhuman intelligence. Just to end, we haven't spoken yet about the Raspberry Pi Foundation. And the foundation is to encourage people, I guess, like the business, to use computing and to provide computers for the developing world. That's it. So it's to get young people all over the world excited about computing and about electronics.

31:26and you've raspberry pi the you go back to the start the diagnosis of the problem was that we had it was there was a missing physical object and the interesting thing is you know that was the dream at first when you know liz and i and other founders were sat down at the start of this we we felt look we'll just make this object we don't need to make any money selling it if we just make the object and get it in the hands of make a thousand and get them in the hands of the write 1 ,000 kids, then mission accomplished, right? Get another 100 applicants a year for Cambridge. It's that, you know, that parochial.

32:00What's fascinating about Raspberry Pi is, of course, partly because of the volumes, and partly because we found a way to make it and e-count a little profit in each unit. We were able to fund the foundation's work. And the foundation has then gone on to do, it's a more traditional educational charity. It trains teachers, it runs after school clubs, it creates teacher training material, it works with governments on curriculum reform. And the money that it raised, and we've given, I think we'd returned about 50 million, it's on the order of 50 million dollars, 50 million pounds, 50 million dollars, to the foundation of dividends before the IPO.

32:35They then raised another 180 million. They've been extremely successful at leveraging that flow of money from us with philanthropic money with individual and corporate philanthropy. And this is now an organization, if you think we were thinking, how do we get another 100 applicants a year to computer science at Cambridge? This is now an organization that can do that work, but can do it at scale all around Britain, and increasingly does enormous amounts of work in India, does enormous amounts of work in the United States, does enormous amounts of work in sub-Saharan Africa. So it's really taking those lessons that they prototyped in that first 10 or 15 years of their existence and then deploying them globally at scale.

33:16And we still do, back to the AI point, we believe this is incredibly important. There is nothing about these tools that frees people from the obligation to be able to think. And that's what, really, that's why I want my children to learn to program computers, is it teaches you to think. You can't effectively program a computer in 6502 assembly language or C or Python or by prompting clawed code. You can't effectively program. There's all different ways of programming a computer. And you can't do that actively if you can't think.

33:59Nothing tests my patience quite like sitting on hold, listening to the same four bars of music on repeat. It turns out all that time adds up. The average person spends over 40 days listening to that dreaded hold music. 40 days? And this episode is sponsored by Parloa, the AI platform built to make that a thing of the past. Parloa's AI agents are available 24 hours a day, 7 days a week. Fluent in any language and smart enough to remember every customer across every interaction. Because no one should have to explain themselves twice. From payment processing to roadside assistance, flight rebookings to appointment scheduling, Parloa delivers the kind of customer experience that turns a one-time buyer into a customer for life.

34:47See Parloa's AI agents in action at parloa.com. That's P-A-R-L-O-A dot com.

35:00so it's interesting don't you think to hear from someone who is not as yeah go ai it's gonna change the world yeah yeah he's not ai pilled which is you know what katie this is why we have this podcast because most tech podcasts there's just people like in the middle of silicon valley everybody's you know reading from the same hymn sheet and so it's nice when you have somebody like i think there is a real value in being far away and being in a different environment and seeing these from the outside and hearing from those perspectives of like yeah i've been doing technology for my whole life and this is a really amazing tool the end i was great hearing him talk about those cycles and looking at AI from the 1950s as someone who not only has worked in tech his whole life, but also been an academic.

35:58So has seen those cycles. And you do get a sense of hearing these guys, especially in court, like, you know, they all love science fiction growing up. They were all computer nerds growing up. And you do feel like they are kind of living out their childhood fantasies and really leaning into it in a way that you're like, how much of this is actually real? Like, are we approaching the singularity? Are we all going to be enslaved by the machines? Or as he says, like, no, these are just really powerful tools. And this is, you know, kind of part of the evolution of technology. I just think it's great to hear somebody who's a bit more got his feet on the ground.

36:34And again, nobody knows what the actual truth will be. But I think it's great to hear from him. It's also really interesting to hear about Raspberry Pi. It is such a quintessentially British, very quirky company, I always think. I mean, how fascinating to have the foundation as such an integral part of it, to have a hardware computer company that's making computers in South Wales. And it's such an interesting concept to be making these very cheap, easy to use computers with an educational mission as well. I hadn't realized it was the third computing company in the world. No, it's amazing. I just go back and listen to exactly what he said there.

37:19But yeah, that was fascinating. Yeah. So anyway, great. What an episode. Covered a lot, as always. There's a lot to cover. Well, I don't leave your phone for a moment. 30 minute window. I have to wait for the all caps email from the court. get you know drop everything it's drop everything into that car off you go exactly exactly um look forward to hearing all about it yeah so who knows this time next week we could have a verdict i'm not sure how long it'll take but obviously we will bring you all of the details all the sorted interesting uh details of this really bizarre case but that is it i think for this week's episode if you are enjoying the show drop us a line please at techpod at thetimes.co.uk that is techpod at thetimes.co.uk and we would love to hear your thoughts on today's discussion what's been going on in court have you got a raspberry pie quirky British businesses whatever it is exactly until then see you next week see you next week bye bye you