← Unholy: Two Jews on the News
Unholy: Two Jews on the News
Trump on the Brink of War and Work in the AI Age - with Barak Ravid and Noreena Hertz
01:03:48
Watch us on Youtube: https://youtu.be/Wsz-rgYJeTgCatch Jonathan on tour in Australia As the Middle East enters another moment of acute tension — with one of Trump’s advisers saying there is a “90% chance we see kinetic action in the next few weeks” — Yonit speaks with Barak Ravid, diplomatic correspondent for Axios and Channel 12, about what is happening now and what Donald Trump may be weighing as events continue to unfold. From geopolitics to technology, Yonit is then joined by Noreena Hertz — academic, economist, and bestselling author — for a conversation about the future of jobs in the age of AI, why women may be particularly vulnerable to unemployment, and how these technological shifts are exacerbating antisemitism. Plus: a sports edition of Mensch and Chutzpah — crossing borders and disciplines, from American and European arenas to stories that run from basketball courts to bobsleigh tracks.
Loading summary
Transcript96 lines
- [00:03]
A
The American public is still under the impression of the Maduro operation. US Sends its elite commandos snatch Maduro, bring him to court in New York. This is not the case with Iran. With Iran, we're looking at a big, wide sustained military campaign. We're talking about a war, not an operation. And we're talking about the biggest US Military intervention in the Middle east the since the war in Iraq.
- [00:42]
B
It's unholy. I'm Unit Levy in Tel Aviv as we are all here in Israel in a holding pattern waiting for a decision to be made by President Trump. And it does seem like that decision to attack Iran is near. This week we will be talking to diplomatic correspondent Barack Ravid, who on Wednesday reported Trump is moving closer to major war with Iran. That was the voice you heard at the top of the show. Later, we might need a breather or at least a break to be worried about something else. Economist, academic bestselling author Narina Hertz will talk about the future of jobs under AI and why this technology exacerbates and propagates antisemitism and what could be done about that. Now, as you may have noticed, Jonathan is away. He's on a book tour and a speaking tour in Australia. If you happen to be in that part of the world, it's always nice to meet him in person. So do check out his schedule in our show notes. We'll also tell you that a bonus episode for unholy subscribers is up. Journalist and author Mati Friedman will be joining me for a conversation I truly recommend. It's a key to deciphering Israeli society and Israeli politics, usually overlooked by outside observers. The divide between Ashkenazim and Mizrahim, Ashkenazi Jews and Sephardi Jews. It's hard to understand Israel without understanding that key component. And Mati is a perfect companion for that conversation. One note on this week before we continue, all manner of news from Israel as usual, Nabarak riots and Trump yet again demanding a pardon for net Netanyahu, the official launch of the peace board for Gaza. But what resonated with me and what I wanted to share with you was the interview aired last Friday with Arbel Yehud and Ariel Cunho. Arbel and Ariel were high school sweethearts kidnapped together from the house they shared in Kibbutz Nyiroz on October 7. She was released after 483 days, January 2025. He was released after 738 days just recently in October 2025. They were held separately and in solit and for the first time they sat down for a joint interview. Arbel had said in that interview that she wasn't able to watch stories of other survivors, though what changed for her was when former hostage Romi Gunnen revealed recently that she had been sexually assaulted repeatedly by terrorists before being transferred underground with other captives. Arbel had said about listening to Romy describing her experience, she said this. I went through this from the beginning to the end of my captivity. Just let that sink in for a moment. 483 days. Albert had also said that she tried to take her own life a few times in captivity because of what she went through. But she stopped attempting to do that after seeing her face on a poster in Hostage Square in Tel Aviv, realizing that people were fighting for her. You can't not look at this brave young woman without thinking about international women organizations around the world. All those who stayed silent, or worse, were skeptical about what Israeli women went through, who demanded evidence and details and context, who treated these women differently simply due to their place of birth, as if it was in any way their fault. It took 54 days after October 7th for the United Nations Agency for Women's Rights to meekly say, in light of the undeniable evidence piling up, that, quote, israeli women are worthy of safety, end quote. That was their response. We need to remember all of those heartbreaking facts. We also need to think of the heartwarming fact that the protest worked and that Arbel and Ariel are home and that Israelis donated millions within 24 hours to help these two high school sweethearts, now reunited to try and build back their lives. With everything that has been going on this week, this is a thought and an interview that stayed with me throughout these recent days. And now let's take a breath and try and understand what is happening in our region at this very moment. Barack Ravid is a global affairs correspondent for Axios. He's also the correspondent for Israel's Channel 12 in Washington and a foreign policy analyst for CNN. He's the author of Trump's Peace, which covers President Trump' first term. Barack, we're so glad to have you on unholy this week.
- [05:03]
A
It's great to be here again.
- [05:04]
B
Yunit, let's begin with a disclaimer. It's 2pm Israel time, 7am Washington time, and 3:30pm Tehran time. Anything we say in this conversation is right for these minutes only. We have no responsibility for what happens after.
- [05:18]
A
For these seconds, for these seconds, especially because in, I think two hours from now, from the time that we're recording this President Trump is going to give speech at the beginning of the Board of Peace meeting in Washington where he might speak about the war with Iran. So we really cannot be certain about anything that's going to happen in the next few days.
- [05:47]
B
So it's a topsy turvy world, and that's an understatement. But it does seem when we listen to your reporting in the last 24 hours, that the question marks around whether or not Trump will attack in Iran are turning into exclamation marks, and that he has reportedly at least decided to attack Iran, and that this, this might be larger than, than people realize. This, this is quoting from your reporting. Is that indeed where we are at this, at this point in time?
- [06:19]
A
I, I'm not. I don't think he decided, okay, but he's very close to giving the order. Meaning, there were talks in Geneva earlier this week between the US And Iran, mediated by the Omani foreign Minister. The Americans were asking the Iranians to come. Ask the Iranians to come to this meeting with something tangible. They hope the Iranians will come to the meeting with something tangible. The Iranians did not come to the meeting with something tangible. Therefore, the bottom line of that meeting was, okay, go back to Tehran and get back to us within two weeks with a real offer, a real proposal written down on paper, something we can look at and tell to the President of the United States, this is where the Iranians are. And I think that this means that we are already in a stage where the Iranians did not meet the expectations of the US Whether President Trump will indeed wait the whole two weeks and see what the Iranians come back with, I think that is a real open question. Some people who are briefed on the talks think that he will, others think he won't. And I spoke to a very senior US Official yesterday about those talks, and he referred to the talks in Geneva as a nothing burger. And I think it means a lot because the Americans hoped, really hoped that the Iranians will come with something that will allow the diplomatic track to develop, to make progress, to make real progress, and it did not happen. On the other hand, on the ground, the US Is sending more and more forces to the region. First and foremost, the USS Gerald Ford aircraft carrier and its strike group, they're now still in the Atlantic Ocean. They're gonna enter the Mediterranean in a few days. They will reach the eastern med in, I think, a week or so, maybe a bit more. And this will be, I think this is the main US Asset everyone are waiting for in the Meantime, dozens of fighter jets, air defense systems were sent to the region. There are about a dozen destroyers already in the region, one aircraft carrier already in the region. So the US Has a lot of forces. It's waiting for the Ford because this is more massive firepower because the Ford doesn't come only with the Ford itself and the fighter jets that are on it. It comes with another three destroyers and other warships. So I think that when you look what's going, when you look at what's going on in the negotiation room, which is not a lot, and you look at what is going on on the ground with the military buildup, which is a lot, so you look at that and you know that President Trump is pretty fed up. That's what I hear from his advisors, that he's pretty fed up with this whole process. It's hard not to reach the conclusions that we are very, very close to him giving the order.
- [09:51]
B
Yes. And last time that was in June when Caroline Levitt did say there are two weeks for negotiations. We remember quite clearly that it took a few days and then the President, the decision was to strike. As said, you don't think he still made the decision. If he does decide, will this be a one and done or will this be a larger, longer operation than we actually realize, Perhaps even longer than the 12 Day War in June?
- [10:18]
A
So yes, yes and yes. It is not one and done. This is not. You know, one of the things that I realized in the last few days is that the American public is still under the impression of the Maduro operation. US Sends its elite commandos to Venezuela. They land with helicopters, use all sorts of special weapons, snatch Maduro, bring him to court in New York. Okay, very clean thing. This is not the case with Iran. This is not what we're looking, looking at with Iran. We're looking at a big, wide, sustained military campaign. We're talking about a war, not an operation. And we're talking about the biggest US Military intervention in the Middle east since the war in Iraq, since 2003. There are more U.S. assets in the region right now, more U.S. fighter jets, more U.S. tankers, more U.S. intelligence planes than any other point in time since 2003. So this is very meaningful. And you don't send that amount of hardware to the region if you are not willing to use it. And if you're not preparing for a multi week military campaign.
- [11:45]
B
You know, the interesting thing you wrote in your report is that there hasn't been sort of wide debate, not in Congress, not in the American political or public sphere. About the ramifications of this, I would say there hasn't been a wide debate in Israel either. When we look at the cooperation between Israel and the United States, is this going to be a joint again? If it happens, is it going to be a joint operation the way that the Trump administration perceives it?
- [12:10]
A
Definitely. You know, the 12 Day War, its last three days were a joint military campaign. When the US joined the war, by the way, it started being a joint military campaign a day or two before the US Conducted its strike on the nuclear facilities. When the US Started refueling in the air, Israeli jets, US Military, US Air Force tankers started refueling in the air, Israeli jets that made their way to Iran for strikes. So if that was the case during the 12 Day War, this time around, it is going to be much more meaningful, much more significant. This is a joint operation. It's going to be much wider than the 12 Day War. And the 12 Day War was pretty big and the 12 Day War was mostly Israel. So now when you take the U.S. military buildup and you add the Israeli enforcements that are going to join, you understand that this is going to be massive, massive firepower directed at the Iranians.
- [13:23]
B
Yes. And we should perhaps think of the massive firepower then directed at Israel. Definitely. The Israeli people are pretty trepidatious as
- [13:31]
A
we speak now, not only directed at Israel, I think that it is clear that this time around the Iranians are going to use their short range missiles much more than they did during the 12 day war, and they're going to use it against US bases in the region, meaning in what's left in Iraq, uae, in Bahrain, in Saudi Arabia, all of in Qatar. All those bases are under real threat of being attacked. Bahrain, for example, with the Fifth Fleet headquarters there, this is the number one target for the Iranians. And I heard just in the last few days from several officials that they think that the first Iranian response is going to be a massive artillery and short range missile attack on this base of the Fifth Fleet in Bahrain. This is very close to Iran. You can fire artillery, long range artillery shells from Iran to Bahrain. So I think those are all things that we're going to see.
- [14:38]
B
Yeah, I mean, you're very right to point out it's definitely not, I'm sitting from my perspective in Tel Aviv, it's definitely not going to target only Israel, but also American targets here in the region. I was really interested in something that Carolyn Levitt was asked yesterday. One of the reporters said to her, why does the President think there is a need to Attack the nuclear facilities that he himself said were obliterated. Of course, he did say that, famously after the June attack. And I wonder how much this is actually, and we know it is an argument within the Republican Party whether to do this or not. How much does this do these voices still infiltrate the president's thinking as we speak now, because we're talking about something that's very likely to happen. Are there still people who are trying to stop him?
- [15:22]
A
Definitely. A lot of people around the president don't think it's a good idea to go to war with Iran right now. A lot of people around the president think that there's no real justification to do it from a US national security standpoint, which for the US the main issue is the nuclear program. The debate sort of got stuck on this juvenile discussion of whether it was obliterated or wasn't it obliterated. The Iranian nuclear program got hit very hard. It has been rolled back years. Therefore, at least what I hear from US Officials, they don't think that there's any imminent threat when it comes to the Iranian nuclear program. And this is why a lot of US Officials are saying, okay, if that's the case, what are we doing here? Is this a military campaign that is aimed at regime change? Because this is definitely what it looks like at the moment. And a lot of people, especially in the MAGA movement, are saying, didn't we say that we're not going to do all that? Didn't we say that the war in Iraq was a disaster, that we don't want to do a rerun? Because at the moment, a lot of people in the MAGA movement are seriously concerned about a rerun and that the US Was gonna get bogged down in the Middle east and that this will determine how the next three years of the Trump presidency gonna look like.
- [17:01]
B
You know, it's so interesting because when you try to square this, these plans, right, for a sort of multiple week of many weeks of a campaign and maybe even regime change, it doesn't really work with the narrative of Donald Trump being the man who, as you said today is launching, you know, the peace board initiative in Washington. He likes peace and he likes deals. This is a line that was said by Jake Sullivan. He definitely doesn't seem to be the person who plans ahead for weeks to come. It's very hard to square these two together.
- [17:31]
A
It's hard to square together, but it's also not hard because what Donald Trump. The argument that I heard from Donald Trump when I spoke to him several times in recent months was that who thinks that the 12 Day War is what allowed him to reach the deal in Gaza to end the war in Gaza? Because Iran has been weakened, its nuclear program was rolled back. It had to deal with its own domestic issues. It had less ability to support its proxies, Hamas had less backing. And therefore the other Arab countries, the mediators, Egypt, Qatar and Turkey had more influence on Hamas. And Iran couldn't be the spoiler that it's been for many, many years. So that's Donald Trump's argument for why it was reasonable to go on the 12 day war from a US standpoint. And I think that he still holds this position. He still thinks so. And he thinks that if you as the more Iran is weakened, the more the current regime is weakened, the easier it will be for the US to push forward peace agreements between Israel and its Arab neighbors. And I don't think this is a completely crazy or weird idea. I think it makes complete sense. The problem is that when you enter such a war, you really don't know how you can end it, how long it's going to take and whether it will lead to the point that you want to go or it's going to completely undermine it.
- [19:17]
B
BARACK I can't say I'm calmer after the conversation, but I am smarter. So that's something. I really appreciate this conversation. Thank you so much.
- [19:25]
A
Barack thank you. Yunit.
- [19:32]
B
I don't know about you, but I find that that the way to deal with anxiety could be occasionally to be very anxious about something else. So let's talk about something else that is in the Middle east for a change. Now we wanted for a long time on Unholy to take a deeper look at a topic that in recent years obviously has moved from business and technology pages to the very top of the news agenda, artificial intelligence, which is rapidly becoming central to all of our lives. And it seems that in the near future whoever cannot adopt to it or at least understand what it is capable of will be left behind. It's almost pointless, perhaps impossible, to try and stick with the news agenda here. Right. The latest on AI. Remember how two weeks ago everyone was talking about the social network for AI agents, but really things are moving so fast, whatever was true days ago is irrelevant today. So we chose to take the bird's eye view of this and we want to hear from an expert on how in her words, humanity is sleepwalking into a future we are not prepared for. Narina Hertz is an economist, an academic and a best selling author. She's a professor at University College London, specifically at the Policy Lab where she leads on AI. She also sits on the board of Mattel Warner Music Group and Work Human. And for me personally, the living proof that you should always surround yourself with friends who are smarter than you. Norina, thank you so much for being on Unholy. It's such a pleasure.
- [20:58]
C
I'm so delighted to be on. It's taken us a bit of time, but I am here and I'm glad to be here.
- [21:05]
B
We're thrilled, I'm really thrilled about that. I mean, look, AI obviously a huge ocean. So we wanted to kind of anchor our conversation in three or four main boxes. We'll just tell our listeners. So talking about jobs, talking about Gen Z and how they kind of treat AI I think also be talking about the fear of misinformation and disinformation and of course try to give us a little bit of optimism as well. These are all areas that you have been researching. I want to start with jobs, if I may. And I mean it's been already, I think the sort of penny dropped on the fact that whether years ago we would have thought that it would be the blue collar jobs that AI would be coming from for. We actually realize now that it's the white collar jobs. Right. The lawyers, the accountants, the doctors. What does that mean for a society where, you know, people have been studying and training for years and suddenly their jobs can disappear?
- [22:02]
C
Yeah, and this is now moving from the realm of the hypothetical to something that we're actually seeing happen. And I think that's important to stress. So if we look at the latest US job data, for example, we're starting to see real evidence that especially when it comes to entry level jobs, companies are just hiring less of the lawyers, the young lawyers, less of the young accountants, less of the young financial analysts, the kind of jobs that young people used to, until really recently go to university and really prize as the top jobs are already being diminished and degraded by AI and the estimates are pretty scary. I mean, I was speaking last week to somebody very senior at one of the big AI frontier firms, Anthropic, and he said to me that he believes that jobs are going to be wiped out. White collar jobs, we're going to be talking about 30% of white collar jobs at least wiped out within the next two years. I also was speaking to somebody from DeepMind last week, another senior person there, and he had a slightly different story. He thought it wasn't going to be just so voluminous so fast, but he also was really concerned, especially about entry level jobs. He thought, you know, maybe it'd be some professions first, like coders, ones where AI could most obviously replicate the job. But happening fast so raises huge questions for society. How will people be able to afford to buy anything? How will they be able to afford their rent? How will they be able to afford to eat? How will people if they're not working and even if you solve those issues, what will people do? How will they have meaning? How will they have belonging? How will they have status? And how angry might this group of educated white collar people be when they realize that that contract that they kind of had implicitly with the state, we're going to study hard and then we will work and we'll get good jobs and then we'll pay our taxes and then we'll get looked after. Suddenly is overnight, it seems in danger of being completely shredded up.
- [24:28]
B
Yeah, you use that word overnight. And when you talk about in two years, of course this is, you know, this is very dire. If it happens within two years, 30% of these jobs we didn't say anything about even high tech jobs. Right. People who are talking about will the job I have right now still exist in even nine or 10 months? Is there a case to be made on the economics level? Is there a case to be made to say you're going to have to give people basic income even? I think in our discussion you mentioned the fact that thinking about the fact that the people from these AI companies are going to make a lot of money, they should take part in that kind of basic income. Does that make sense? I mean we're looking at this really a bird's eye view, but that is what society is going to be having to deal with very soon.
- [25:13]
C
I mean it feels inevitable that there will have to be some sort of mass distribution of money by the state to people. If we do have these levels of unemployment that we could be very reasonably looking at given the forecasts. I think the question is, is that enough to be thinking about? Because even if people don't have to worry about having enough money to pay their rent or eat or pay their mortgage, we are back to that purpose of meaning status, belonging and basic income is not going to answer that. So I think there is a question of if this is the sort of scenario that unfolds, is it that people are just given money or do they have to do something for the money? Do governments need to be think about having massive volunteer like schemes, but instead of it being voluntary, you're paid for it. So for visiting elderly people, for picking up litter, do we need massive public arts programs? Which is what happened, of course, after the Great Depression in America when the New Deal brought about a massive civic arts program where people were hired to paint, to sculpt, to create. Is that what governments need to be thinking about? And. And who's going to pay for this? That's a good question.
- [26:45]
B
Yeah.
- [26:45]
C
I mean, who is going to pay for this? Should it be the biggest AI frontier firms, you know, should they be required? Is this their license to operate? They're going to be making a lot of money. Should they be funding this? Should governments be thinking creatively, perhaps setting up a sovereign wealth fund in which they take stakes in all the new AI firms that spring up in the hope that some of those are going to become very rich and the government will become rich as a result? Should it be companies who are deploying AI the fastest? Should they have to pay some sort of tax which is then used to help people who are being displaced? I mean, lots and lots to be thought about. What really worries me, and I spend a lot of time with governments across the world, what really worries me is that governments are not thinking about this. They're barely thinking about this, and they definitely are not committing significant resources to thinking about this. And yet we are talking about a very, very short time frame when everything could be upended in ways we only starting to realize.
- [27:58]
B
I'm curious why they're not thinking. Is it because our reality, the political reality, the social network reality, is such that the politician can only allow himself or herself to think about the next tweet and they can't think long term? Or is it because they don't understand it in depth? I mean, what would be the reason for not thinking about this long term?
- [28:17]
C
So until recently, they could probably use the argument this is going to be somebody else's problem, you know, not the problem for when we're in office, and so they can push it into the long grass? And I think that's probably to some degree what happens. I think also they're just not enough people within government who typically are on top of this big AI transformation and really understand what's happening with the level of granularity that you need. There's also the fact that the last few years have been tough for governments across the world. We're living in this very fractured world. We've been living through economic downturn, high inflation, high interest rates. And so governments have had a lot of other stuff on their plate. In Israel, you've got all, of course, the defense issues to be thinking about too. So a lot that governments are grappling with. And it's also because I think people often don't want to focus on the bad, on the bad news. They don't want to focus on the downside risk. And maybe, maybe it's easier and happier for politicians to instead of thinking about all the bad things that could happen if I really took off in the way that it looks like it will, instead, just focus instead on the very positive narrative that AI companies themselves are spinning, that AI is going to deliver this world of great abundance, of great productivity, and companies who are thinking hard about how to deliver growth and maybe thinking, let's just focus on that and not think about the downside.
- [29:57]
B
One of the more interesting things I think of your research is to point out, I haven't seen people do this to point out the issue of gender, that when you talk about job loss, you focus on the fact that it will probably be and sadly be women who would pay that price more than men. Could you go into that a little?
- [30:16]
C
Yes. So the estimates are pretty stark and bleak. It's estimated that three times as many women in high income countries will lose their jobs due to AI than men. And to unpack that a bit, it's useful to think of the 1980s as the parallel, because that was when the PC came on people's desks. And what we saw was a really clear gendered impact of this new technological transformation. Because which jobs were replaced first? Data entry clerks, secretaries. These were the people who lost their jobs pretty quickly when the PC arrived. And how the story unfolded was not a good story for women. Because what we see when women were tracked over the subsequent decades, what we saw was that many of these women either never got a job again, never re entered the workforce, or if they did re entered it at a much lower level of pay, they went into care work, for example, which was much lower paid. So there is a real danger that we will see this unfold again. Especially as if you look at what the top 40 jobs most likely to be replaced by AI are, these are jobs like translators, proofreaders, marketers, customer service operators. These jobs are disproportionately majority female. So you've got history showing us this has happened before. We've got the fact that the job's most likely to be displaced by AI. First, the majority of these are predominantly female. And then we've also got the fact that Whereas in the 80s, what we saw was that there was One group of women who actually fared better, and that was the women who adapted fastest to the technology and learned the new technological skills the fastest. What we're seeing play out right now is that women are lagging behind men when it comes to this. So women and men are using ChatGPT, for example, in their personal lives pretty much the same. But when it comes to the workplace, we're seeing a real divide emerge, a chasm emerge between the extent to which we're women are using it and the extent to which men are. It's partly because women say that they're less confident using the technology. It's partly because they're more risk averse and that, you know, in some ways they have right to be cautious about this new technology. But it's also because it turns out that companies are investing less in upskilling their female scarf when it comes to AI than their male staff. So, you know, something for women to really be thinking about, are we upskilling ourselves enough? You know, presuming there will be at least a transitional period when there will still be something for people, humans to do, you know, are we making sure that we are skilled? You know, that is serious. Then at least let me be one of the ones being hired. And companies need to think, are they inadvertently, perhaps prioritizing their male staff rather than their female stuff, which of course in most countries would be in breach of equality legislation. So something to think about there.
- [33:46]
B
Yeah, that's really interesting. I'm thinking we're talking in like very short time spans now, but I'm thinking if one was to ask you, what should my children study? I mean, we were talking about this the other day. My eldest, who is 12, Yonatani wants to, he's studying code. He wants to be a coder. Great. Assume. I don't want to break his heart, but I assume that AI can do a lot of what he's, you know, studying to. We should say he's 12. Right. But the point is that what should children go into? I mean, should they study law, accounting, medicine or should they be doing something completely different?
- [34:21]
C
I think it's wonderful that Yonatan, age 12, is learning how to code and teaching himself to code. And I think, I think that's wonderful because it shows that he's curious.
- [34:31]
B
Yeah.
- [34:31]
C
And that he's interested in the world and that he's trying to figure it out. Those skills we're going to need regardless, like whatever happens, kids are going to need and should be developing their powers of curiosity, their powers of kind. Of being interested and their ability to learn, adapt, discover. Coding itself I would say is not a good profession.
- [34:54]
B
Not the best investment I know. Okay.
- [34:56]
C
No, when I was speaking to Anthropic, they tell me that whereas a year and a half ago they were using Claude for about 5% of their code, six months ago that had risen to 50% and as of now it's about 90%. So 90% of their codes is not being written by humans but by AI. So Jonathan doesn't want to be a coder, but. But I think kids today do need to become AI ninjas. They do need to really understand this new technology and actually they can be assets then in companies when they join, when older people may just not be up to speed, frankly. So that can be a positive. At the same time, I do think kids and young people, all of us in fact need to be ring fencing AI free time because it's so easy to default to AI and young people are increasingly doing this. 50% of 13 to 17 year olds are now regularly using AI
- [35:59]
B
and we
- [36:00]
C
have to make sure that our brains don't atrophy, you know, that this doesn't become the Google Maps situation at scale where it's not just that we can't read maps anymore, but we just can't think anymore. So ring fencing AI free time I think will be key for kids to have an advantage because then they will still be able to do things that AI can't, although these may be diminishing, but also have the confidence to be able to assess the AI, judge it, challenge it when need be, which will remain, I think, really important skills. I think kids today also young people need to be really developing a resilience because we really don't know how this is going to unfold. And so they have to be able to cope with change, adapt to change, pivot as well, and retain optimism. I mean their parents may be less optimistic, but I think we definitely want our kids to remain optimistic at this time.
- [37:14]
B
You know, one of the interesting areas of your research has been into sort of young Gen Z and how they deal with AI. And one of the interesting things is just the sort of emotional level. It's not only knowledge that they're asking from the AI, it's can we say this? Even emotional, you know, friendship, things like that. That is, that's very weird for our generation when we look at it.
- [37:40]
C
Yes. So they're using it in a few really interesting ways. They are using it for their studies for sure, not perhaps always in the most obvious ways. They're not all using it to write their essays, although a lot are. Some of them, you know, proudly tell me we're not using it to write our essays, but we are using it to outline and structure our essays, which of course we know is the hardest bit. So they are using it for their academic work for sure. But what I found most interesting was the extent to which they were using it to navigate and pass their social interactions. So, for example, you know, I had an argument with my friend, this is what I said, this is what she said, who was right, or this is what my boyfriend messaged me, what does it mean? Or they're using it for things like, as one of them said to me, for my opinion, for having an opinion on anything they explained to me, for example, they used it, should I give money to the homeless person on the street? They were using it for when their parents sent them messages. They were using it to work out how should I reply so that it sounds a particular way. These were 18 to 21 year olds I was interviewing and doing focus groups with. What was particularly disturbing was that they said that they were worried that their younger siblings were using it in very similar ways. So their younger siblings were using it to ask the kind of advice that they would have asked of their parents or of them. They were now asking of ChatGPT. Their younger siblings were asking it the advice when they were trying to navigate bullying at school or kind of difficult situations rather than an adult or a trusted person. So it was the extent of it. I mean, some of them were using it for every single decision they were making. One young boy was using it for everything, literally, what shall I have for breakfast? What shall I wear today? Should I go to the library or should I study in my room? And what they said to me, and I found this quite heartbreaking, was that they said that they now trust the AI more than they trust their own judgment because the AI gets it right, whereas they're not sure they do. And that skill that we develop when we are, when we were teenagers and we were young, that awkwardness, that getting it wrong, that making a mistake and
- [40:26]
B
the learning, falling down and getting up
- [40:28]
C
again, right, in general, that's something this generation risk not doing, with the result that they won't be exercising those psychological muscles that are actually really essential for when you are in the world and have to navigate face to face in person interactions and try and read a room and work out what people are really saying and navigate arguments. The danger is they won't ever have done it because they will have always defaulted to the AI, which they believe is neutral, which they believe is all knowing, which they believe is smart. When of course it's not necessarily any of those things.
- [41:13]
B
Is there a way to look at it perhaps in an optimistic way, just to say, okay, this is a very, very turbulent world and so many uncertainties and sometimes you are lonely and maybe this is good, you have something to talk to when you're very desperate. Is there any sort of way to look at this less worryingly? Because the instinct is of course to say this is really worrying, that there's a whole generation who's not going to know how to live in this world emotionally. But is there a way to look at it in a different way? Perspective?
- [41:46]
C
I think there is. And in the work that I have done on AI companions specifically, I think from an individual perspective, you can imagine real benefits, you know, for the child who is struggling with making friends and doesn't really have friends. And we know that young people are really lonely today. Gen Z and Gen Alpha are the loneliest generations in history. So for that person, having an AI companion, you know, could make them feel much less lonely, much less alone. But again, I think there is a danger there, dangerous that they interact with these companions who we know are sycophantic, who we know will always make you feel great, that they won't have to learn how the give and take of real conversations. And also from a societal perspective, what kind of a society will we be creating if people don't feel a need to be there for each other because they know that actually, you know what? It's fine. My lonely child, my lonely parent, my lonely friend. I don't need to be there for them because they can sit in a room speaking, to chat instead. What kind of a society would that be?
- [43:09]
B
Yeah, it's also the chicken and the egg, right? You don't know if people are very lonely and Gen Z, as you say, because of the technology to begin with. So that explains a lot about where we are if we're not worried enough. Let's talk about the issue of misinformation and disinformation in the world. I mean, we should mention you're part of the Jewish community in the uk, your great grandfathers, in fact, the Chief Rabbi of the uk. And when you think of, you know, things like biases about anti Semitism, even about the fact that Israel is, is going into an election year and what that means, the fear of, of misinformation and disinformation, where do you Find yourself on, you know, on the scale of 1 to 10, how, how worried are you of all this that is going on?
- [43:50]
C
Oh, don't make me put it on a scale because then I'm just, my audience is going to be, our listeners are going to be, viewers are going to be very depressed. But I am worried. I'm worried because if we just even look at last year, there were a number of elections. Ireland, Moldova, Poland, South Korea, Ecuador, where we saw AI playing a part in disinformation. So you talked about elections in Israel. And when we saw this last year, we saw fake AI, CNN reports, falsely implicating politicians in scandals. We saw deep fakes in South Korea of a politic, you know, looking like they were in prison attire in a cell. We saw social media farms, you know, with hostile actors kind of enabling social media posts to be ever more engaging, seeming, but essentially be propaganda. So, you know, this is real. We saw it last year. AI's just getting better at this stuff, just getting faster, just getting better, easier for bad actors to deploy. If you've been looking at some of these latest iterations of video, for example, I mean, it's almost getting to the point where it is literally impossible to discern the fake from the true. So really worrying when it comes to potential for misinformation and disinformation when it comes to elections. As Jews, we need to be thinking about the impact AI is already having on antisemitism because it's already having a significant and very disturbing impact, really, for three reasons. First, because if you think about how generative AI works, you have these big models which are trained on text. And a lot of the text these models have been trained on is bad when it comes to truth veracity about Jews, about Israel, about Zionism, but also material that actively promulgates antisemitic tropes and conspiracy theories. A lot of LLM training material is Wikipedia or Reddit forums. So, you know, of course we can't trust that the output these models are going to generate is going to be output that has been screened for truth. So I think that's one problem. I think another problem is that people receive the information. When you put in a search on ChatGPT or such like, you get an answer. It's not like when we search things on Google, at least we used to get a page and then many pages of answers. So at least perhaps part of you realized that there were maybe contesting truths and contesting narratives and other information out there. These LLMs, they're putting out information in the form of one fact, one fact that seems, you know, authoritative. So if you're not somebody who does question things, you're more likely to believe whatever it says. And if what it's saying is anti Semitic or unfair or anti Zionist or not saying something true about Jews or about Israel, you're less likely to see that there is a corpus of competing truths and narratives out there. There's also the fact that it's just so much easier now for bad actors, whether we're talking jihadists or anti Semites of the right or the left or or nation states, to generate very persuasive content that seems like it's real. So, you know, there are videos out there of, you know, Jews with kipot dancing outside the when the Twin Towers are falling down, for example, there are videos out there, Jews with fangs kind of drinking blood. And unfortunately, to a mass largely uneducated, all too gullible public, these videos can seem real. And these videos are then of course, amplified, shared, etc. So you've got the fact that you've got this kind of. The material itself is trained, the output is trained on bad, untrue material. You've got the fact that it's just so much easier to create anti Semitic content. And then you've got, and then you've got the fact that the owners of these platforms really don't seem to care. I mean, there was a piece of research that the American Defense League did recently where they tested all the major models. They put in 50 prompts, see whether they would generate antisemitic videos. And over 40% of the cases they did, over 40% they did with Grok Elon Musk's platform being the worst on that front. It's not that they can't do anything. If we think about something like extreme child porn, for example, you know, it's not that it doesn't exist at all on social media, but by and large these companies have devoted enough resources, enough thinking to stop a mass of that sort of content being unleashed online. But they don't seem to have the will, they don't seem to care enough on the whole. They're definitely not investing enough money into trying to fix the problem. And they need to be, they need to be partly responsible and largely responsible. But governments also could do much more. I mean, they could be saying to these AI platforms and also of course to the social media platforms who are then the vectors for this, this AI disinformation and anti Semitism, they could be saying to them you are responsible for everything that is on your platforms and if this stuff is there, you're going to get levied with a huge fine. They could be saying we're not going to use your platforms when we're doing government procurement, which is of course, you know, a big way these companies make money. We won't use you until you've sorted this out. But again, governments are not, not doing enough.
- [51:12]
B
We've been having this conversation and I wonder if after saying all this, there is something to be said about this technology that is making the world better. I mean, things are usually when this kind of technology hits planet Earth, like other technologies, like the industrial revolution, there's always going to be, it's going to be a mixed bag. What are the good aspects of this
- [51:35]
C
that you see first? Good shit does happen.
- [51:41]
B
I'm printing that on a T shirt. Okay, good.
- [51:44]
C
I mean, it does. And whilst I do think the downside risk is real and I do think we need to wake up to this and I do think we are in danger of sleepwalking into a future that we are not prepared for, this isn't a Luddite anti technology position because when it comes to medicine, I think AI is going to bring about almost inconceivable gains. I mean, we're already seeing when it comes to screening, for example, AI being better than humans at reading scans, at spotting cancers earlier, at not missing problems. When it comes to personalized medicine, AI is going to be able to have a huge impact in making sure that the treatments we're offered are the treatments most likely to be able to serve us. When it comes to drug discovery or drug trials, we're already seeing that AI can have a really significant impact in speeding this up. So medicine is one area where I think unequivocally and when I speak to cancer researchers, when I speak to Alzheimer's researchers, when I speak to leading medics, this is an area where unambiguously, I think there will be a net positive. I think the other area where I am optimistic about the role that AI can play is around what we might think of as democratization of expertise and democratization of support. So, you know, it used to be very expensive. If you needed legal advice, if you wanted to see a fancy accountant, if you needed financial advice, if you needed specialist medical advice, this was all something that for many people was out of reach or hard to access. I think expertise. And there are caveats because at the moment these models do make things up, a lot up. They do hallucinate. But as they get better. And presuming this problem is solved, there is a future where this sort of expertise is available for people en masse, not just the rich. And I think that is something that's encouraging. I think there's also a democratization of support that is exciting. You know, think of something like tutoring for kids. I mean, this is really overwhelmingly a middle class privilege at the moment. You need to have a private tutor on top of your classroom teaching. But again, this is something where AI could really democratize opportunity. Every child could have a tutor in their pocket who can help them. And the studies on AI tutors, provided that they're well designed, are actually super encouraging. There was a study done recently at Harvard which showed that AI tutors were actually better when well designed than the kind of education people were getting in the classroom. So I think that's another area where we can be excited that people and entrepreneurs, I mean everyone, if you have that gene, if you have that in your DNA, you won't have to raise lots of money before you can have a marketing team and a sales team and a social media team and financial analysts. You'll be able to do this yourself. So I do think there is opportunity here for sure, and I do think there are things to be optimistic about. But at the same time, I don't think we're planning for the future that could be unfolding. I don't think we're putting in the guardrails to deal with the problems that AI can engender. Misinformation, not only, of course, when it comes to elections, but as Jews when it comes to misinformation and disinformation and anti Semitism, which we see AI playing a real role in exacerbating. Where are the guardrails? Where are governments insisting on it? Why are we not putting enough pressure on the AI frontier firms to be managing to be doing something, to be using all their brains and smarts and money to address not only how they're going to get richer, but also the downside, societal risks.
- [56:28]
B
Norina Hertz, it was a pleasure to talk to you on Unholy. Thank you so much for this conversation.
- [56:35]
C
Thank you so much for having me on.
- [56:42]
B
We have arrived at our awards section of the show. Now Jonathan is away, so obviously a frowning face and a small chutzpah award making its way to him. But this will not deter us. We have asked our loyal listeners to send us their suggestions for this week's Chutzpah and Mensch nominees. And here to help me sift through all this in her debut on the podcast, our inimitable producer, Michal Porat. Hi, Michal.
- [57:10]
C
Hello.
- [57:10]
D
How are you?
- [57:11]
B
Good. We should tell our listeners not only our producer, at times kindergarten teacher, perhaps psychologist shepherding her unruly co hosts of Unholy, also Fashion Police. I mean, you have a lot of titles on this.
- [57:25]
D
I have a lot on my plate.
- [57:27]
B
That's right. We usually have our coverage conversations in Hebrew, but.
- [57:30]
D
Yeah, yeah, we have to. I have to say it's. It's weird, but we'll get through it. So let's. Let's jump in.
- [57:38]
B
It's weird, but we'll get through it. As a motto for life, I think that's every day.
- [57:43]
D
So we turn to our, as you said, loyal listeners on Instagram. If you're not following us yet, then
- [57:52]
B
do you with your life.
- [57:53]
D
Yeah, that's right. So we asked for a mensch and chutzpah nominees. Should we start with the chutzpah?
- [57:59]
C
Maybe?
- [58:00]
B
I think that's the right order of things. Okay.
- [58:02]
D
So did you watch the All Star game this week?
- [58:05]
B
A little bit. A little bit of it. Okay. It was late at night is real time. But yes, we made history. Yeah.
- [58:11]
D
I hope you were cheering for the world team because this year it was USA versus the world, which is, I think is a great cast. Yeah. So we had our own player over there, Danny Avdia.
- [58:27]
B
He's for the first time.
- [58:29]
D
Yeah. Probably the most successful Israeli in the NBA. And this time he was elected for this all world team with really the best players from all over the world that are playing in the NBA. But our chutzpah comes from the audience. Did you see. Did you spot Spike Lee in the audience?
- [58:50]
B
Yes, we should say Spike Lee arriving with a sort of keffiyeh pattern on his sweater and a Palestinian flag kind of for his bag. I mean, I don't know if this was more chutzpah for that or chutzpah for the way he was dressed. That's your decision. That's your jurisdiction to do fashion or just sports chutzpah. But, you know, I understand how people think that even in sports you need to have that insert politics. It's just, you know, could you give us just the. It's been a tough year here. Tough couple of years. Could you get us. Give us.
- [59:23]
D
Let's just chill for one night. Yeah, definitely. So obviously seeing Danny playing on the court and, you know, with the Israeli flag and then Spike Lee with his political statement is something that our listeners found deserves to be called chutzpah I tend to agree, but we have another nominee. So maybe Spike Lee won't be winning this week.
- [59:49]
B
Puts award another nominee. In the same vein, we should say
- [59:52]
D
a much colder setting for this one. I'm not going to ask you if you watch the Winter Olympics. Thank you for that, by the way.
- [60:00]
B
It is open in our house. Like there are other people in the household that are.
- [60:05]
D
Some of my best friends watch the Winter olympics.
- [60:07]
B
Exactly.
- [60:10]
C
So A.J.
- [60:10]
D
edelman, brother of Alex, friend of the pod. He was on our show in New York last year.
- [60:18]
C
So he's. You.
- [60:18]
D
You have to say I. I don't know even how to pronounce this sport. I mean, bobsledding, it's not something we're used to in Israel.
- [60:26]
B
Bobsledding. I mean, yes. It's not incredibly Jewish, is it? But.
- [60:29]
D
Yeah, yeah.
- [60:30]
B
But he has a team and we made, by the way, just to make note, he was our Mensch Award recipient last week. And now we're turning this whole story into the Chutzpah Award. I like this. It's like Phil Dunphy said it, Modern Family. When you have lemonades, turn them into lemons. So that's what we're doing this episode. So our nominee is the Swiss commentator who while the bobsledding team for the Israeli Olympics was actually a time when they were competing. This is what he said. Sorry to quote this, but our listeners were emotional about this. So let's give the whole quote. He said Adam Edelman, who self defines as a Zionist, posted several messages, messages on social media in favor of genocide in Gaza. This is while the team is competing. He did not post anything we should say. Adam Edelman, supporting the genocide in Gaza. He is a Zionist and he is competing for the Israeli team. And that was enough for a Swiss commentator in a sporting event to talk while the Israeli team was competing.
- [61:28]
D
Yeah, really taking advantage of your ability to talk to large audiences and just, just, you know, spread your own opinion. That has nothing to do with what people are seeing on, on the screen or the sports itself. Okay, so do you want to choose? It's a tough competition for Speckley, I have to say. I think that's, I mean, in fashion terms and, and sports, it's really.
- [61:52]
B
I agree. This is one award he doesn't have on his shelf, so why not give it to him? We have a very themed mansion Chutzpah a section this week. Because our, our Mensch award also has to do with everything we've just talked about.
- [62:04]
D
That's right. So we have a suggestion from Emily Einhorn from Instagram. And she nominated LeBron James. I think first time appearing in this. Yeah. First time nominee doesn't have to be the last time.
- [62:20]
B
LeBron.
- [62:21]
D
So let's hear what he said about Danny Abdia and more surprisingly, about Israel.
- [62:26]
E
I mean, he is an All Star. He's playing, you know, exceptional, you know, basketball. I've never been there. If I have fans over there, then, you know, I hope you have been following my career. I hope I inspire people over there to not only want to be great in sports, but I want you to be better in general in life. So hopefully someday I can make it over there. Like I said, I've never been over there, but I've heard nothing but great things and.
- [62:49]
B
Very nice, Very nice. He went on to say, I think he has many fans here, which is true, and that he might come visit. We're very glad about that. We'll welcome him anytime he wants to come. Not recommended in August. It's very hot. You're very welcome.
- [63:04]
D
Yeah. And maybe another group match for all the Israelis that flew from all over the globe, really, to watch Danny Abdia in the All Stars. First Israeli ever, as we said, playing the All Star game. I think that's really, you know, just a very, very Israeli moment, flying from all over the world to see one person play basketball. So ending on a nice note here indeed.
- [63:29]
B
So, Michael, very big thank you for helping us out in the award section this week and of course, for everything else you do on Unholy. And we might be back with emergency episodes. We hope not, but if not, we will see you next Friday.
- [63:43]
D
Bye.
- [63:44]
C
Bye.