BBC World News AI Decoded 30April2026 Stephanie Hare

Añadido: 2026-05-11

[00:00:03]Well, now it's time for AI decoded.

[00:00:12]Well, welcome to AI decoded and this week saw the opening of a legal clash that could shape the future of artificial intelligence. Tech titans Elon Musk and Sam Altman are currently locked in a dispute over the true purpose of ChatGPT creator OpenAI. Well, Musk is suing fellow OpenAI co-founder Altman alleging that he misled him after the organization shifted away from its nonprofit roots towards a for-profit model. Meanwhile, OpenAI says that Musk is motivated by jealousy and regret for walking away from the company back in 2018.

[00:00:48]Well, BBC News online reports that the Tesla and SpaceX billionaire was the first witness to give evidence telling the court that he wanted OpenAI to remain a not-for-profit because of concerns that artificial general intelligence AI could surpass human intelligence and it might otherwise fall into the wrong hands. Meanwhile, The Guardian reports tense exchanges in court with Elon Musk repeatedly sparring with OpenAI's lawyers and prompting several interventions from the judge who at times urged him to give clear yes or no answers. Well, Musk accused OpenAI's counsel of asking misleading and deliberately tricky questions. Well, also on tonight's show, we look at how a growing number of students who are switching courses at university in favor of those who teach human skills in order to AI-proof their future careers. In a recent study, CNBC says that Gen Z adults worry about AI's impact on their cognitive abilities and almost half say that the risks of using AI in the workplace outweigh the possible benefits. Well, to guide us through everything AI decoded, we are joined by Stephanie Hare, who is with us tonight.

[00:02:02]But before we hear from Stephanie and all of her thoughts on that showdown between Elon Musk and Sam Altman, we have this report from our North America technology correspondent Lily Jamali, who is outside the court in Oakland, California.

[00:02:17]Elon Musk was combative under cross-examination by OpenAI's lawyers, who he accused of asking overly complicated questions. "Your questions are not simple," he told the lawyer at one point, adding, "They're designed to trick me." OpenAI tried to paint Elon Musk as out for control of the company until he left in 2018. Musk acknowledged he wanted initial control over OpenAI, but said he intended for that to change quickly as more investors got involved.

[00:02:45]OpenAI's lawyer also tried to poke holes in Musk's contention that he cares deeply about developing the technology safely. He briefly raised Musk's relationship with US President Donald Trump, who has adopted for less regulation of AI, not more.

[00:03:00]OpenAI is seeking to show that Musk is suing in part because he's upset that the company went on to succeed without him. Musk returns to the stand on Thursday for more cross-examination. Mr. Altman is also expected to take the stand at some point during the month-long trial.

[00:03:18]Well, Stephanie Hare, as I mentioned, joining me in the studio tonight.

[00:03:21]Stephanie, really good to see you here on BBC News. So, we've got this big legal clash between Altman and Elon Musk. Just bring us up-to-date with, I guess, the key arguments in all of this and where we are in this case.

[00:03:35]>> So, we're right at the very beginning.

[00:03:36]The trial opened yesterday to great fanfare, and we think it's going to last about 3 weeks. So, what is going to be decided? Many, many years ago, before any of us had long heard of ChatGPT, in a little town in Silicon Valley, a group of people got together and formed a non-profit and called it Open AI, which of course has created ChatGPT that many of us use today.

[00:03:58]This was supposed to be, as Elon Musk says, a charity, and the idea was to build AI and think about AI in a way that would protect humanity, so we did not get a Terminator-style outcome, some of the scarier sci-fi options. So, the people who formed this took it really seriously, and Musk was one of them, and he invested a lot of money in it. He feels that Sam Altman, who was the CEO of Open AI, was one of the the founders as well, is betraying the original mission. He's made it from a non-profit to profit. He also thinks that he's making very dangerous technology, and he would like to see him ousted as CEO, and to get some of the money back, but put towards the charitable side of the company. Now, is Musk a honest broker in this discussion? We will have to discuss that as well, because of course Elon Musk has built his own technology. So, if he's really worried about AI destroying humanity, why is he himself actively building it? As you point to, I mean, the accusations in all of this from Sam Sam Altman has been, "Look, you left the company, you're jealous over all of this." Elon Musk, as you have been pointing out, is saying, "Well, look, I have very legitimate concerns about AI falling into the wrong hands, essentially what this could mean for humanity." As you point out there, Stephanie, this case will start to look at how legitimate those concerns are when it comes to Mr. Musk, but more broadly, how big are those concerns when it comes to AI? Because it is a pretty scary prospect. It is a scary prospect, and I think that's what's so maddening for all of us who study AI and all of us who just have to live with it. Is on the one hand, we're told AI is really dangerous and it could destroy humanity. We're told that by really serious people, by the way, the people who are the so-called godfathers of this technology, some of the biggest investors in it, some of the top scientists. We're also told by everybody that it's going to lead to incredible prosperity. People won't need to work. Sam Altman has said we will enter into a world that is post money. We won't even need money anymore because there'll be just so much abundance. So, between these two extremes, we somehow have to find reality.

[00:06:04]Yeah, and that will be of course the big question. I mean, and whether of course people will legislate off the back of that as well, you know, depending what we see happening with jobs. And on that point, really interesting to see that we're looking at a growing number of students turning up at university now, you know, either choosing their courses with an eye to what's happening with AI or kind of watching this unfold slightly alarmed saying, "Look, I want to switch courses because how do I AI proof my career?" Tell us more about those findings. So, there's a number of factors that play there. I think first we have to acknowledge the fact that the cost of getting a university education in the United States is almost unaffordable for a huge number of people. That wasn't always the case. So, something's changed in the American funding model that might not necessarily be true in other countries where you can still get an affordable education. So, is this an American phenomenon because of that kind of a factor or are we seeing this with young people around the world? So, that's the first thing we'd want to test. But the second thing is this idea of AI proofing your future.

[00:07:08]And what does that mean? It used to be, at least in the United States, that if you got a computer science or computer engineering degree, you could write your own ticket. That has changed in the past couple of years. And we're trying to figure out the reasons why for that.

[00:07:21]Partly it's AI, but it could also be some other things, changes in how they do recruitment. There was massive over hiring during the pandemic, so a lot of companies are actually unwinding that.

[00:07:30]It's not very productive or efficient.

[00:07:32]So, again, very, very complicated. One of the strategies of young people is doing things that that AI just simply could not do. So, let's take an example of being like a hairdresser or a florist, right? Or a doctor or a nurse, but even then you hear people in medicine who are worried about that. I think we have to push back against that though, because AI often replaces tasks, not necessarily entire jobs. There's a lot of money going into AI, but the productivity statistics are not great at the moment. It has not demonstrated its value. And if I had a dollar for every time someone's asked me, "Well, what on earth should I do next in terms of a job?" And you say there that, you know, this can be with regards to to tasks as opposed to just kind of, you know, the thinking that goes into all of these jobs. But I I thought it was interesting that also in this study Gen Z are concerned about the impact from AI on their cognitive abilities. You know, is that kind of just hearsay what people are experiencing or is there kind of concrete evidence of the fact that relying on these tools too heavily is having an ability an impact on our ability to to think critically? It's absolutely having an impact on people's skills. So, people who use AI for writing, so that might just be your emails and maybe that's okay, but if you're using it in report writing, they're finding that those people, if you ask them to explain their work, can't because they didn't really do it, right? And you see that also with coding. So, everybody is worried about losing skills, losing knowledge, and it's great to say, "Oh, we'll just prioritize constructive thinking, critical thinking." But you have to earn that the hard way. There's no shortcuts for critical thinking. So, using AI to do your homework for you or hack your workload, I understand that everybody's under a lot of pressure, but you might actually be cheating yourself in the end. I mean, the message has been in recent weeks from the former Prime Minister Rishi Sunak, you know, that companies are relying on this, particularly when it comes to those entry-level jobs, that it's showing up.

[00:09:33]He says it's starting to show up in the employment statistics. If you look at people who have recently graduated instead, some consultancies and so on are saying, "Look, for these more simple tasks, when people start to climb the ladder, we can use AI instead." And I think that brings us to the question, are politicians starting to grapple with this, legislate? Because if you've got a lot of young people who are struggling to start their careers, that's a problem. I think politicians could push companies to ask themselves, "Do you need to redefine what we mean by entry-level?" Maybe entry-level doesn't need to be stuff that you're doing automatically and by rote. And a great example of that would be people going into law, for instance. So, they were having to do a lot of just going over contracts and doing research, looking things up. That's how you get the expertise that you would then use to be able to check something that's AI-generated and know where there's mistakes. You have to build that muscle memory, if you will, the hard way. Well, maybe we need to redefine what we mean by being an entry-level lawyer to make sure that you retain the knowledge and skill set, but still get the benefit of using AI for stuff that you don't want to be spending lots of time on. So, I think it's making us rethink work and rethink value and rethink our jobs.

[00:10:44]And what about those companies who might be very tempted, essentially, to use AI where they can, they think, to save money on the workforce? You know, one of these ideas that have been floated by Rishi Sunak was, "Okay, well, let's see taxation for these companies." You know, could there potentially have to be measures on those companies that rely too heavily on AI to replace members of the workforce? I mean, I think we want to be really careful about that, because then we're getting government really tinkering in capitalism. And I know it sounds wonderful on paper, but historically, it tends to not work super well. So, I think what I would say is, first of all, particularly the big American technology companies already don't pay enough tax correctly. Like, we don't need to name and shame them. We all know who they are. So, first, just tax them properly first and get your tax harmonization policies working across the board, so they can't go hide in Ireland or the the Netherlands, right?

[00:11:33]And and do that. It's like you could you could do that without having to like bring AI up at all, and you'd already do a lot of good. Second, I think what you need to do really is be again be thinking about what do we want the future of work to be. If we just use AI to automate things so that we can then cut head count, right? As a cost-saving measure, that has a short-term win, but you can only take it so far. Once you've trimmed the so-called fat and you're running lean, that's it. That does not answer your question for growth. It does not answer your question for innovation. What new products are you coming up with? What new services are you coming up with?

[00:12:05]That's where you go back to human talent. It's going to be humans who will still come up with those things and humans who will buy them and sell to one another. So, we have to think really holistically about what is an economy and what role do we want AI to play in it. We should be investing more in people, not in less. You know, that's really interesting. I mean, do you think that when you've got something shiny and new like AI, people look at it and think, "Oh my goodness, I can outsource all of these tasks." But then when they actually look at what it really achieves and they drill into that, they might think, "Oh, maybe not necessarily what I was hoping for. Still need a human." I just think people are actually really conventional in their thinking about what they could do with AI. They're using like some old capitalist playbook, which is like, "What can we do to cut costs?" And then that's head count, right? And they're very excited about automation cuz it can run 24/7 and it doesn't need sick leave or holiday pay and it won't talk back to you.

[00:12:55]It has other problems, by the way, but it at least doesn't do what those human human qualities would be.

[00:13:00]The real value is going to be using AI to do things that humans currently cannot do.

[00:13:06]Yeah, indeed. There's no law or regulation for that. That's just human innovation.

[00:13:11]Well, we're really going to hope that we can still have this conversation on programs like The Context. Stephanie, thank you very much for being with us.

[00:13:18]Pleasure.

[00:13:20]Well, if you'd like to watch this week's full AI decoded episode as well as previous ones, you You always scan [music] the QR code on your screen now, which will take you to BBC AI decoded playlist on YouTube. Of course, you can catch up on all the latest episodes there. You can also watch the show on BBC iPlayer.