Charles Fain Lehman, Rafael Mangual, Tal Fortgang, and Carolyn Gorman discuss Columbia University’s deal with the Trump administration, the White House’s AI action plan, and the technology’s usefulness in tasks at work.
Finally, a reason to check your email.
Sign up for our free newsletter today.
Audio Transcript
Charles Fain Lehman: Welcome back to the City Journal Podcast. I’m your host Charles Fain Lehman, Senior Editor of City Journal. Joining me on the panel today are sometime host Rafael Mangual, who also works on all things crime at the Manhattan Institute, Tal Fortgang, who works on all things domestic extremism, legal, whatever at the Manhattan Institute, and Carolyn Gorman, who works on all things mental health at the Manhattan Institute. Thanks everybody as always for being on.
I want to take us into the news the past couple of days, coming out of right here in the city of New York, actually where I am for the day, specifically at our old friend Columbia University, which has finally reached a deal with the Trump administration, settling a variety of complaints. And then also I think on Tuesday, they suspended or expelled 70 students who were involved in the takeover of a library that I believe we talked about on the show previously. So this seems like a big about-face. Carolyn, you were going to summarize the details of the agreement for us.
Carolyn D. Gorman: Yeah, so like you said, Charles, this agreement between Columbia and the Trump administration was just reached yesterday. Columbia is going to pay $221 million in penalties for allegedly failing to protect students from anti-Semitic harassment. And I say allegedly failing to protect students because the agreement does not require admission of fault. But in return for paying these penalties and meeting a few other demands, Columbia is going to get hundreds of millions of dollars in federal research funds that were cut off back in March, including most of the grants that had been terminated from the National Institutes of Health and the Department of Health and Human Services.
So, a few other things aside from these penalties that Columbia is going to pay, they’ve got to change their disciplinary review, which is what we saw on Tuesday. They have agreed to review their Middle East curriculum to make sure the curriculum is balanced. They’ll appoint a new faculty to the institutes or excuse me, to the university’s Institute for Israel and Jewish studies. They’ll add an administrator to serve as a liaison to students facing anti-Semitism. And the university also promises to end programs that promote race based, outcomes like admissions, which the university was doing illegally after the Supreme Court’s decision ending affirmative action. All these measures are meant to be enforced by an appointed federal monitor.
Charles Fain Lehman: Okay, so thank you for that summary. I want to dive right into the details here, and I’ll frame the conversation by saying there are parts of this that I like and parts of this that I don’t like. I like the cracking down on campus activism. I am, and this is, think, very live conversation. There are parts of this that basically go, we need to be more aware of campus anti-Semitism. And the way that we’re going to do this is like, you know, incorporating greater oversight, finding new ways to incorporate, essentially incorporate Jews into the DEI bureaucracy. And then I’m like, this does not sound great to me. As somebody skeptical, you know, who is Jewish and is skeptical of the DEI bureaucracy, but I wonder, I wonder what people make of the overall package. And then also, you know, do we take Columbia seriously? It seems like they just kicked out a bunch of students. What’s up with that? What do we, what do we make?
Rafael Mangual: So I think there are a couple of things to point out here and I think you should start with the political nature of this dispute. And I think that it is unquestionable that this is a 100 percent decisive victory for the Trump administration. Not only does this agreement with Columbia University represent an important disciplinary action or accompany an important disciplinary action for exactly the kinds of students that critics of the university have been complaining about for some time, but it actually comes with real accountability mechanisms like massive fines, the appointment of an independent monitor, et cetera, that Columbia is going to be held to. So on top of all that, though, the university’s leadership in its statement about the agreement, and this is, I think, a really sort of important note here, said in no uncertain terms that the agreement safeguards and protects the institution’s values and, more importantly, its independence. And that’s important because one of the most potent critiques of the Trump administration kind of trying to hold universities accountable in this way is that it was going to interfere with academic freedom and Columbia University just kind of cut itself off from that argument in its statement. So I think on all fronts, this is a win, certainly a lot more than anyone would have expected to get even a couple months ago. So I think it’s absolutely a good place to start.
Tal Fortgang: Yeah, there are parts of this that smack of merely expanding DEI, right? More monitoring, more reporting, more individuals and institutions who will make Jews and Israelis feel more welcome. And as ever, the devil is in the details there as to what that entails. I tend to be a little bit skeptical of things like expanding Israel studies and Jewish studies because those are often departments, much like American studies, that really have an implied “anti” at the beginning, right? They’re about the critical assessment, not in the good sense of being critical, but in the sense of being uncritically critical of their subjects.
Then there’s the matter of admissions, right, which as Carolyn mentioned correctly, Columbia has now committed to doing what they were already committed by law to doing, right? Which is to emphasize colorblind meritocracy in admissions. The open question is whether admissions will have some mechanism for accounting for the kinds of radicals and lunatics who have who had to be expelled and suspended recently because Columbia was just full of totally unhinged young activists who were courted, right? These admissions departments selected for certain characteristics aside from identity characteristics so that they could have campuses full of these people. Have we addressed that issue? That’s not clear.
Rafael Mangual: Well, look, I mean, I think there’s certainly been at least an attempt to address that issue when you have, you know, I forget how many students it was, but something like 70 students, you know, either expelled or suspended in one fell swoop. I think that does send a clear message, and as activists as these kids are, they’re all very concerned about their future and their future earnings. And I do think that this kind of disciplinary action is going to function as a deterrent for a good portion of these. In some sense, it doesn’t really matter if they continue to court people with these kind of characteristics, if those people are sufficiently deterred from the kind of misbehavior that we care about.
Charles Fain Lehman: Yeah, I mean, I think, right, so there are two components there. One is that I think that’s probably right is the university has finally said if you commit actual crimes on campus, there will be consequences. We won’t arrest you. I mean, we’ll arrest you. There’ll be no prosecution to be sure. You won’t actually be held legally accountable for crimes like trespassing. But eventually you might be suspended months and months later, which like is probably an effective deterrent. Yeah. Oh, please. I was about to get there.
Tal Fortgang: Can I just… Can I play cynic though? We don’t I don’t know. Do we know how many of these students have actually been expelled? Because we…
Charles Fain Lehman: No, it’s 70 suspended or expelled.
Tal Fortgang: Because these suspensions are very, very often temporary measures to get the media or government or some critics off the university’s back. Remember, no less than Mahmoud Khalil, the head of CUAD, the terrorist-sympathizing rabble rouser, is now doing his media circuit.
Charles Fain Lehman: An American hero, an advocate of free speech.
Tal Fortgang: Yeah, exactly, the principled free speech anti-war peace activist, who simply loves everyone and just wants the dying to stop and distributes Hamas literature on campus. He had been suspended. He was previously suspended, I think, at least once. But that doesn’t mean anything if it’s temporary. An expulsion goes on your permanent academic record. An expulsion can ruin your life, which is, frankly, the proper outcome for people who break laws, who really, they do more than break them, they flout them, right? They laugh in the face of laws in addition to school rules.
Charles Fain Lehman: I do very briefly… The other point on this front, I think there’s at least some argument for Ralph’s position vis-a-vis deterrence, which is just like lot of Columbia students, a lot of students in major universities are like, they have a highly elastic demand with regards to consequences. Like you scare them a little bit. These are all strivers. They want to get jobs at Goldman Sachs. And look, Goldman Sachs does kind of want to hire Mahmoud Khalil, but like eventually they’ll come to the idea that hiring these people is a bad idea.
That to me is, even if they moved to a little bit of the margin, I still think it makes a difference in behavior. The big question to me is just like, in some sense, the big question to me is, how effective is the anti-Semitism bludgeon? I have no problem saying Columbia has an anti-Semitism problem. They do. There are a lot of crazy anti-Semites on campus. That’s real problem. But that is not the root issue, which is that these universities are propounding insane ideologies and instilling them in American children. One of the side effects of this is that they also become raving anti-Semites. But like, you know, they also hate white people. They also hate Asian people. They also want socialism. You know, not to sound like crazy person, but to me the question is basically just like, why I’m, part of why I’m concerned about the Title VII and Title VI stuff, you know, the IHRA, the, you know, expanding the definition of anti-Semitism is that it reinforces, can re-entrench, let off easy organizations that are sort of at the core of the rot in universities that Americans are concerned about not just because they happen to propound anti-Semitism but because they are corrupt from the ground up. So that’s my, you know, that’s my question. I’m going make sure we get Carolyn in here, also Ralph, then Carolyn and I also, yeah, Ralph has been trying to get in.
Carolyn D. Gorman: Yeah, no, Charles, just to sort of back that though, I think like when we think about whether or not these measures are going to work, like that type of kind of ideological bias and corruption is just so sort of deeply ingrained in so many places that it’s hard to see this ending up really making a change without sort of instead just kind of making conflict maybe less apparent on campuses. So there really needs to be like a thoughtful and consistent effort and really cultural change if we think that something is going to come out of this long term.
Rafael Mangual: Yeah, look, I just want to say, I think we need to manage our expectations about the potential for affecting cultural change on a university campus like Columbia is, right? I Columbia University, yes, absolutely has an anti-Semitism problem, absolutely has an anti-America problem, absolutely has an anti-white problem, an anti-establishment problem. But those problems have existed for much, much longer than we’ve been having this conversation. More recently, I think, the kind of post-October 7 climate sort of brought to the fore, problems that were already deeply ingrained in that university, I don’t think we’re going to necessarily change that. But I do think it’s a win if we create the kind of circumstances that make educational life possible and more suitable for people who maybe don’t agree and don’t fit into that milieu. And that, I think, is the most important thing that’s actually getting done here, right?
I mean, I don’t really care that much if I’m sitting next to people who hate me or who hate my ideology or who hate my country. As long as you leave me alone and I can go about my business, we’re all going to be just fine. It’s when you start blocking my access to my library. It’s when you start making noise outside my dorm room and I can’t sleep or study in peace. It’s when you start forming human chains and blocking me from getting from one part of campus to the other. That’s the stuff that needs to stop. As long as that happens, I think that the strongly minded, conservative, Jewish, white and Asian students who have absolutely been on the receiving end of the sort of university establishments ire or for generations really, I think they’re going to be just fine. And that’s the starting point here. The other thing is that the independent monitor is going to be in place, right? I mean, that can be a source for future action, right? To the extent that the independent monitor is saying like, hey, these are the problems that we’re seeing and documenting. That’s going to set up the case for future action. think that’s important.
Carolyn D. Gorman: One of the ways that this could be useful is these suspensions like to your point, point Charles, I think students do kind of have, a lot of elasticity in terms of like what they will pursue bad behavior for. And if this sort of disrupts the momentum that sort of was just growing and growing, I think that that is a good thing because the momentum disruption hopefully will just sort of actually quiet people down and make this less sort of socially attractive to young people.
Charles Fain Lehman: Yeah, I mean, to me, the core question is basically just like my brief for the Trump administration strategy in higher education generally is that there is a there’s a consummate failure of deterrence on the part of Republican administrations, right? Colleges were doing often flagrantly illegal things. There were no consequences imposed. They were engaged in open discrimination. We’ve done a bunch of reporting on this, right? It’s not just like it’s not just discrimination admissions, which was like ambiguously permissible. They were probably crossing the line, but whatever, prior to Students Fair Admissions. There was discrimination in hiring. There was discrimination in the allocation of scholarships. It was all of this stuff. They were just engaged in flagrantly illegal behavior. And the federal government under Republicans just of let them do it. And the Trump administration is like, OK, we’re going to whack you some. We’re going to try to reestablish returns.
And the key question to me is, will that actually happen? Is making example of Columbia, making example of Harvard going to communicate to other universities that we, you know, the next time a Republican is in office, they should expect similar stuff and so they don’t want to open themselves up to liability? Or is Columbia going to like give them an opportunity, know, create a playbook for routing around that kind of threat? That to me is the core question.
Carolyn D. Gorman: Well, one thing we should think about in terms of how readily schools might be willing to come to an agreement is think about the amount of the fine. Like $221 million, that might sound like a lot of money, but it’s worth putting that in context of the size of university endowments. Harvard’s endowment is something like $53 billion. Yale’s is something like 41. Even really small schools have like over a billion dollar endowments. And so if you think that that is that paying some fine is going to sort of get the administration off your back, I think it seems plausible that other universities would take this path.
Rafael Mangual: Yeah, look, let me just say too, it’s not just entirely the fine, right? I know they say that they’re not admitting to any wrongdoing, but just like, you know, any other kind of confidential settlement, right? You can still hold that over and say, you know, in a congressional hearing, you know, isn’t it true that you settled, you know, $220 million lawsuit, you know, with the federal government that was investigating, you know, discrimination on your campus, right? And they’re going to have to answer that question in the affirmative. And that’s always going to be a bad PR look irrespective of whether they technically admit wrongdoing or not. And so that alone, I think, actually can have a deterrent effect. And that’s something that I think a lot of other universities will want to avoid, even if the actual dollar amount doesn’t upend their budget.
Tal Fortgang: Yeah, this, the process and the costs of the process comes with a possibly an equal and opposite cost of the politicization of this issue, which I don’t mean in a bad way. I don’t think it’s wrong for Republicans to hold our institutions of higher education to like basic standards of social value, right? Like to be useful to Americans, to do something good for Americans rather than undermine and corrupt Americans with conspiracy theories. But to the extent that Republicans do that, Democrats are pushing twice as hard in the opposite direction saying this is all a McCarthyist witch hunt, right? Totally politicized issue. And then when Democrats are back in power, we should expect all of this to just swing back and for the federal government to become especially solicitous of universities trying to make back whatever they’ve lost, right? I think this will even out. So I mean, in that sense, Republicans have to continue escalating eternally. So I don’t know what you make of that.
Rafael Mangual: I mean, the one issue I would take with that is that Columbia and its statement says the opposite, right? They’re not attacking this in the way that Harvard did when the federal government first kind of took action against them and said, well, this is a terrible infringement on our independence and we’re going to stand against it, right? Columbia saying actually, no, this agreement protects our independence and it protects our university’s values. And I think that matters. I think that’s actually a really important signal.
You know, yeah, I think that’s worth looking at. And the other thing, too, is that, you know, universities are going to, think, find themselves potentially in the following situation, which is when they’re going to be open to the critique that like, hey, you know, at the end of the day, you can get yourself out of this situation as soon as you decide to by simply cutting yourself off from the faucet of the federal government’s money, which you don’t really need. Right. I mean, you have Hillsdale, you have other models of colleges and universities that do this just fine. And I do think, you know, there’s a big part of the American populace that doesn’t really have any patience for a university with a, you know, multi-billion dollar endowment, you know, taking hundreds of millions of dollars a year from taxpayers, you know, to fund its efforts that it could afford to pay for on its own, much in the same way that, you know, like Oliver Anthony saying, right, if you’re five foot three and you’re 200 pounds, taxes ought not to pay for your bags of fudge rounds. I mean, it’s the same concept, except our welfare queens in this space are Ivy League universities filled with mahogany rooms and 40-foot ceilings and rare book collections valued in the hundreds of millions of dollars. So, yeah.
Charles Fain Lehman: It’s a nice rare book collection. Yeah. Nobody goes in. They are very pretty.
Carolyn D. Gorman: I like the rare book collections.
Charles Fain Lehman: Yeah, I want to take that as an opportunity to take us out. so I’ll ask. I know Carolyn hates this, but I’ll ask everybody to make some predictions, which is, do we expect Columbia to be an outlier is really the question. Which is to say, they have in some sense been more compliant with the administration than other schools. Arguably, one model of this is because they know that they really screwed up and another model of this is because Columbia wanted the excuse to deal with the particularly bad problems they have on campus. Do we expect similar agreements like this? Do we expect similar disciplinary action at other schools, or do we expect Columbia to be an outlier? That’s the question. Tal, I’ll start with you.
Tal Fortgang: Yeah, well, Colombia was kind of the worst offender to begin with, right? It’s been a hotbed of total academic, social, cultural corruption for a long time. It’s been a symbol of that. If this is the settlement that they reach, right, as Carolyn pointed out, not a tremendous amount of money, you know, an independent third party oversight, you know, whose effectiveness is to be seen. I’m confident, but I’m not assured that that’s going to work, then I think schools that had lesser problems, that did not have Columbia’s full-fledged history of encampments and occupations and Middle Eastern studies departments, that you would not believe the so-called scholarship that they produced. Like the equivalent of that for Harvard is like really not that bad. It’s really like a slap on the wrist. So I would expect this to be a model, but calibrated towards each school’s problems.
Charles Fain Lehman: Fair, okay. Carolyn>
Carolyn D. Gorman: Well, you know I don’t like to make predictions, but I mean, I think Tal sort of has a point. Having something that is kind of tailored to each individual school and coming to some agreement based on that doesn’t seem totally outside the realm of possible.
Charles Fain Lehman: Ralph, what do you think?
Rafael Mangual: Yeah, I think this is the first domino in a series to fall. I think you’re going to see other schools come to similar agreements. You know, they may not be as harsh. Their dollar amounts may not be as high, but the subject matter may also not be the same. I’m sure that there are other universities that are offenders on different fronts, including in the admissions space, even post-SFFA. And so, yeah, I think this is going to be a sort of central part of the federal government strategy to hold universities accountable to not just, you know, protecting students against rabid anti-Semitism, but also just ensuring the kind of colorblind equality that our Constitution requires.
Charles Fain Lehman: Yeah, you know, I think I think a big part of this depends on where litigation with Harvard goes, but particularly if the administration gets its act together there, like you get wins in Columbia and Harvard, that I think is how you start to establish deterrence to go back to my framing earlier. It’s just like you have to make clear that the Republican Department of Education, if it continues to exist, we’ll come down hard on all of this stuff. The Department of Justice will probably still exist and they’ll come down hard on it. Um, cool. Let’s, let’s move on from higher education to the thing that’s going to destroy higher education.
Yesterday, the White House dropped a new AI action plan. It takes a distinctively pro-AI posture, advances the regulatory agenda, emphasize the need to compete with China. It’s sort of one big caveat is that it’s going to make sure the AI isn’t woke. This is, I think, this to me is really interesting because it fits into this broader phenomenon of the sort of partisanization… The development, possibly developing partisanization of artificial intelligence is an issue where I think over the past couple of years you have seen, and this is one interpretation, but you’ve seen a shift towards being pro-AI as a more Republican thing and being skeptical of AI as more Democratic thing. I’m not sure that’s totally set in stone. Um, I’ve seen critics of this action plan see an unrestrained pro-deregulatory agenda as a thing that could do extraordinary harm to the public, including ending the world, if you’re a believer in the Eliezer Yudkowsky theory of things. So I want to kick it to the panel. I joked before we started that had a panel full of Luddites here. But what do we make of this? What’s our take on the White House’s approach?
Carolyn D. Gorman: Well, no Luddite here. So the action plan, just for those who haven’t read it, it resets the policy position on AI to be to attain and maintain American global AI dominance. And it’s really framed as foundational to national security interests. There’s kind of three main pillars to the plan. First, to kind of accelerate innovation by getting rid of red tape. Second, to address infrastructure needs to keep pace with innovation. And then third, the plan has sort of recommendations for international security and diplomacy around like things like sharing innovation with our allies.
I got MI fellow Danny Crichton’s take on this action plan this morning briefly. Danny is definitely someone everyone should follow for all things tech and innovation. And his take is that this is a very strong plan that really first and foremost emphasizes American competitive strength. In his view, it’s better to dominate through superior technology and better supply so that the world is running on our technology rather than trying to restrict American exports, which just ultimately hurts American companies. And look, I think this makes a lot of sense to me because whether we like it or not, AI is here and it’s going to shape our world. So we need to prioritize innovation.
Rafael Mangual: I think that’s exactly right. But to me, the thing about the plan that stood out to me the most was its sort of deregulatory emphasis, which I think is actually important because this is something that’s been missing from the federal government’s approach to lots of industries over the last several decades. And it is also interesting to me because it doesn’t seem to quite fit necessarily with the tone of President Trump’s two campaigns, which are sort of very sort of geared toward working class America and against the idea of shipping jobs overseas and replacing workers. There’s no question that AI is going to do just that, right? It is going to take jobs. I think it will also probably create jobs and the administration seems to be making that bet. But to me, the most important thing is that what I don’t see here is the same kind of fear that I think a lot of people would have expected on the regulatory side from an administration like Trump’s. And I think that’s a really positive development and it represents growth and it represents the fact that he’s got the right people around him who can speak to the benefits of deregulation when you’re introducing a new technology like AI into the economy. And I think that’s going to be great.
I mean, we’ve been around long enough to remember all the worries about stories about innovation in different industries and what effects that was going to have. We automated in agriculture and that was going to take people out and I forget what percent of the American population was employed in agriculture and then that dropped off a cliff. But very recently we had one of our lowest unemployment rates ever. The economy will grow, the economy will adjust, I think good things will happen and I think a lot of that is going to be a function of the fact that they’re taking this kind of deregulatory approach. It’s not crazy, it’s not a free for all. It’s not entirely laissez-faire, and there are good national security reasons to be careful on that front. But I do think that it’s open to innovation and that to me is the most important thing.
Carolyn D. Gorman: Look, I think we need to worry about jobs, but like, we still don’t know what this looks like. You know, we should be aware of it. What we really need to do is make ourselves capable of building things like infrastructure, like data centers, because we just currently are not. I mean, like you mentioned, Ralph, this is just so true for everything from housing to energy projects to like public works. We just have so many regulatory roadblocks and legal roadblocks and political roadblocks. All these things make it really hard to build things in the physical world. And that type of stuff takes a long time, but it’s going to be necessary for, really leading on this. So, there are a lot of aspects of AI that I definitely think we should be thoughtful and, you know, if not worried, at least sort of cognizant about, but particularly when it comes to infrastructure, the stuff really takes a line that it’s worth speeding up.
Tal Fortgang: It’s a little difficult for me just kind of as like an AI layperson. I’ve like played around with some of the LLMs, you know, your chat GPT, your Claude, whatever. I’m not that impressed with the quality of writing and I’m sure a lot of people share that experience. But the truth is that we really get barely a surface level view of what artificial intelligence and machine learning actually are, right? Like LLMs are not the full extent of AI and it’s a mistake to conceptualize AI as being just LLMs because if so, I would breathe a sigh of relief. But there’s a lot more going on beneath the surface and I think we need to be in conversation, you know, both like the four of us and sort of culturally with the people who are actually working on the stuff behind the scenes and can explain what its potential is realistically within the next few years.
Aside from the radical humility of taking that position, I’ll just say that the only thing worse than AI, like completely elbowing out the rest of our economy and the knowledge economy in the next few years is Chinese Communist Party control of AI or Chinese dominance in AI, such that it becomes a national security threat. So I think that simply trying to understand clearly what AI’s potentials are in the next few years and then establishing dominance, over China. That’s a perfectly reasonable way to approach it. Beyond that, regulation is kind of just this like hazy word. I don’t even know what regulation would look like at this point.
Charles Fain Lehman: Yeah, I mean, you know, I think in some senses there are two ways to frame the AI problem. One is the sort of economic impacts and I’m persuaded by, you know, a version of the argument that Ralph has laid out, which is that basically you get returns to increasing efficiency for every, for every, you know, lost job, you create additional jobs because of the increased efficiency associated with AI. That’s sort of borne out by the limited research that we currently have, which says that AI seems to improve efficiency without job penalties. That could get bigger. Like we don’t know what the economy scale effects are going to be as of yet, but it’s like, you know, that that seems to be the pattern of innovation historically. I think it’s probably right, by the way, to the point about economic impacts. But do you this gets to an interesting split in the sort of magnet economic coalition, where on the one hand you have people who are sort of, they want to return to the pre-services and manufacturing economy, and there are other people who want to push this forward into this sort of higher tech, higher skilled economy. And this seems like a victory for the latter faction over the former faction. But that’s one way to cognize the threats.
I think to Tal’s point, the other way that I think about this that I think we really aren’t talking about is just the social implications of AI. You know, thinking about this very concretely, people’s substituting of AI for interpersonal connection, the use of AI to generate deepfakes, the use of AI to generate pornography, the use of AI to interrupt an already interrupted social fabric. In some senses, the government can be involved in regulating that. I’m not concerned about that on a principled level. But I think there’s a deeper set of challenges of articulating the moral problems associated with AI that we are particularly ill-suited to handle as a society at this exact moment when they are most pressing. So that’s really where I am concerned, the death of moral philosophy at the time of artificial intelligence. Totally a new sub-stack post title.
Tal Fortgang: That’s good.
Rafael Mangual: No, look, I think the deep fake problem is a real one. And it’s one that gets a treatment in the action plan. That’s worth noting. But I do think this is where my kind of inter-libertarian comes out, because I think the market’s going to provide an answer for this. I do think that there are going be services that you could download onto your computer that’s going to… No, but I do think it’s going to…
Charles Fain Lehman: Markets are going to sell us AI pornography. That’s what they’re going to do. That’s what markets do.
Rafael Mangual: People will be able to have access to services that will immediately tell you if something’s a deep fake or if it’s genuine, that we’ll be able to sort of evaluate these things and help kind of put a lot of people at ease who are worried about the very real problem that deep fakes could present if that sort of solution doesn’t come about, but I think that it will. And as those kinds of solutions develop and the AI world begins to grow, I think that’s going to put a lot of the concerns that people have at ease.
Carolyn D. Gorman: You know, just to go back to one thing you said, Charles, I think a lot of times we worry about how this is going to impact like the lowest income or lowest-educated workers, but there is some research out there that suggests these are individuals that could be helped.
Rafael Mangual: I think they’re going to be the best off, right? Like the plumbers, the construction workers, the people who are working with their hands. I don’t think we’re going to see code make a hammer magically float in the air and drive a nail into a two by four.
Carolyn D. Gorman: But not just for that reason. You basically give all these individuals access to really high-quality things like legal advice or medical advice or the capacity to communicate professionally via email. You know, if you have your own small business and you are not wanting to send all these really nicely worded emails to your clients, you sort of have the capacity to do so now. So, that’s just one aspect that I think maybe goes under-discussed when we talk about implications for lower income workers or lower educated workers.
One thing that we also, I think, need to be thinking really deeply about is not just impacts on the higher education industry, but just impacts on education in general. I mean, we’re asking, I think, maybe the wrong questions about AI in education, not whether we shouldn’t be asking like whether or not we’re using AI to cheat, but whether or not we’re equipping young people with the skills that they’re going to need to use AI and just sort of take in information, consume information. I mean, if 30 percent of American kids can’t read, we probably, at grade level, then we probably have a bigger issue to worry about. Those sort of fundamentals are going to be important.
Charles Fain Lehman: I’ll give Ralph the last thought than one takes out.
Rafael Mangual: Yeah, I mean, on just on the education front, know, this worry about AI, I mean, you know, coming out of law school, I think I have a different perspective on this because I mean, lawyers don’t really have original thoughts when they make arguments, right? We all get on Westlaw or Lexis and we, you know, look up the cases that are related to the ones that we’re working on and we, you know, basically copy and paste arguments, but the skill is in identifying which arguments are going to help your client and which aren’t, and how to order them and how those arguments fit within broader legal frameworks. So there’s still a lot of work that has to be done. You can’t… You know what I mean? I’m just less worried about that prospect than a lot of other people.
Charles Fain Lehman: So I want to take this out and I want to do that by asking a question that just personally interests me and I ask friends about this all the time. How much do you use AI in your current workflow? You know, do you use it daily? Do you use it multiple times a day? Do you use it weekly? What is your current frequency? Carolyn, I feel like you might have the highest answer that three of them are on me.
Carolyn D. Gorman: I use AI every day for work and otherwise. But for work, I use it anytime I’m working with data because it’s a really good forcing mechanism for me to sort of think through and fully conceptualize what I want to do analytically before I start coding something up. It’s easy to sort of get into bad habits when you’re looking at data and it’s interesting when you don’t have sort of like a clear game plan in mind to just do a bunch of stuff and then you’ve spent a lot of time doing a bunch of doing a bunch of different analyses and they don’t actually progress your work. So thinking about what, you know, my prompt is to sort of write the code for me in a good way, actually sort of helps force me to take the time and think through what it is specifically that I want to do. And then it just writes code so fast and it’s, you know, is it the most beautiful or most organized code always? No, but it’s getting good enough.
Charles Fain Lehman: I don’t trust a AI to write my code. I’ll start there. Ralph, how about you?
Rafael Mangual: Yeah, for me it’s probably a couple times a week and it’s much less involved. I usually I use it to kind of sift through the things that I’m thinking about reading and to sort of figure out what’s going to be helpful, what’s not going to be helpful. And I’m not really a quant either, so just having it explained to me, can feed it a regression table and it can sort of break things down for me and links. That sort of stuff is helpful.
But I’m still not entirely sold on the quality. I mean, there have been instances, for example, where I’ve asked it to summarize a paper that I know very well, and it will give me a pretty significant error in the summary that departs from what the paper actually says. And so I end up kind of double checking it anyway. So it’s not quite maximizing my efficiency, but I do think it has that potential. So I’m hopeful that it’ll get better.
Charles Fain Lehman: Fair. Tal?
Tal Fortgang: What’s code? I don’t, I don’t, I’m not familiar with that. Is that, is that a mili- okay, well, I… Chat GPT is useful if you know that there are phenomena, there’s, there are examples of a phenomenon out there and simply Googling keywords is not going to get you exactly what you need, right? Like, chat GPT, give me 10 examples of, universities that failed to suspend students after they occupied a building. I could Google some of those keywords. ChatGPT does a better job of actually identifying the instances, the particular kind of thing that I’m looking for, and cutting out some of the noise. Of course, sometimes it does hallucinate, so always double check, go right to the source, the same caveats as ever, but it’s really a research tool for me.
Charles Fain Lehman: Yeah, I mean I…
Carolyn D. Gorman: Can I just say, I think the best way to sort of learn when it is sort of giving you the wrong answer is just to use it all the time because you sort of then start figuring out how to ask for things in the way that’s going to get you what you want. And if you’re not practicing and asking in different ways, it’s a lot harder to sort of suss out the difference between what’s not really information you should trust and what is.
Charles Fain Lehman: Yeah, I certainly use it all the time for all sorts of things, including recipes. It’s really good for recipes because it’s synthesizing.
Carolyn D. Gorman: Yes!
Tal Fortgang: News from the law, not to distract from Charles’ burgeoning cooking skills, news from the law is that a federal judge in, I think, New Jersey apologized and retracted an opinion after the judge produced an opinion that had some hallucinated quotes and some opinions. It’s actually it’s not totally clear whether this was an AI problem or a bad clerk problem. I think it probably was, but to my knowledge, this is the first time we’ve seen it…
Charles Fain Lehman: It’s the clerk using AI. Yeah.
Tal Fortgang: we’ve seen it from from this side of the bar, right? We’ve seen advocates whose like whose time is quite limited and all that. This is the first time I think we’ve seen, I would say, an ex-clerk using technology in this unfortunate way.
Charles Fain Lehman: Alright, before we go, I want to turn our attention to something a little bit lighter. A frequent panelist, Renu Mukherjee, sent me some recent survey work from our friends at AEI that, among other things, highlights a 13-point decline in the share of Americans who say they have a third place, a coffee shop, bar, park, etc. that they go to regularly versus 2019. So I want to ask, do you have such a place? Where is it? What is your third place experience? Ralpf, you’re nodding. You take lead.
Rafael Mangual: It’s my gym. It’s my gym. I love my gym. I am in there every day. I often work at my gym. They’ve got like a little work space. They’ve got a pool, indoor, outdoor, with like, you know, a bar and restaurant that, you know, so we take our kids on the weekends. It’s, that has absolutely become my third place. Did not spend nearly enough time in the gym prior to the pandemic. That was one of the gifts of sort of working from home was that it allowed me to get back into shape. So, yeah.
Charles Fain Lehman: I’m thinking of joining your gym. We should talk about that offline. Carolyn?
Rafael Mangual: You should. Yes. Yes.
Carolyn D. Gorman: Mine is definitely not a gym. I wish I could get myself to go to the gym. I used to ski every single weekend and going to the same place sort of you see the same people and it was a big part of my life growing up. I live in Texas now so that is sort of a problem but as my fiancé and I are sort of thinking about where we’re going to live long term we really want to be near somewhere that’s know driving distance to a mountain because it’s a good place. It’s a good way to be outside all the time, even in the winter and, um, you know, see the same people hang out, be active. So maybe that’s my gym.
Charles Fain Lehman: Fair, okay. Okay, Tal, how about you?
Tal Fortgang: Well, Carolyn, I know a house in Northern Nevada that’s going to be available for rent in short order. For those watching on video, we’ve blurred out the moving boxes behind me, but, you know, I can hook you up with the right people. Yeah, I go to synagogue every Saturday Shabbat morning. We live in a very small Jewish community. Otherwise, I’d probably be there on Friday night and you know, ideally every day, every morning as well. But yeah, on Shabbat morning, we get together, we pray, we chat, we drink a little. A small but tight-knit community. You meet some really interesting people in non-traditional Jewish communities, and that’s become our third place.
Charles Fain Lehman: This is what I’ve discovered is that the more observant you get, the more heavily they drink during the services, which I think is a virtue. Yeah, my answer, as all my colleagues know, is my local Panera Bread, which is my home away from home. On that note…
That’s about all the time that we have. Thank you, as always, to our panelists. Thank you to our producer, Isabel Redjai. Listeners, if enjoyed this episode, or even if you haven’t, don’t forget to like, subscribe, hit the button, ring the other button, click the third button, do all the buttons in front of you at this time or later on YouTube and all other platforms where you listen to us. Don’t forget to leave us comments, questions, concerns down below. Until next time, you’ve been listening to the City Journal Podcast. We hope you’ll join us again soon.
Photo by CHARLY TRIBALLEAU/AFP via Getty Images
Source link
