Loading summary
Transcript90 lines
- [00:01]
A
Kevin, you are at like the almost one year mark. You're like a full Texan now. You don't have a signature draw,
- [00:07]
B
maybe.
- [00:08]
A
You reek of barbecue. And we can't tell over the interwaves.
- [00:10]
C
You got to get into Jiu Jitsu, Kevin. They've got some great down there.
- [00:15]
A
Jits. We're calling it.
- [00:17]
C
That is what you call it.
- [00:18]
B
You know, I don't know if Jiu Jitsu's on the menu. We're getting into stand up paddleboarding on Town Lake. So when you all visit, I've got a spare and we can, you know, go for a row, hang out.
- [00:29]
A
Oh, I like that. I like that.
- [00:30]
B
I do have a top 10 list now of breakf. So if I reek of anything, it's just having way too many breakfast tacos.
- [00:36]
A
I will say, the one time I got to visit Austin, since starting at Lawfare, I think I reached out to Bobby Chesney, wonderful friend of Lawfair, obviously, Da UT Austin, because I was going to ACL and he had a plot. He very kindly invited me to join a plot, but I was there with a bachelor party with a bunch of friends and I got just too drunk that I felt comfortable interacting with the Dean of UT Law School. I'm not irresponsible, but I'm like with a bunch of guys who like started drinking at 9am and it was like 2 in the afternoon. I'm like. I'm just like, his kids are going be there. I feel like this is not a good impression to be made, so I just skipped it. So I feel bad.
- [01:08]
B
No, no. I will share one thing about Dean Chesney, who is a remarkable human being. In addition to being a excellent scholar, an incredible dean, he plays for a band at something called Chicken Shit Bingo. So you can go see the Dean of Texas Law jamming away with his buddies at Chicken Shit Bingo in Austin. And this is my sales pitch to all those students who are trying to decide which school to come to. Come to Austin. Come see your dean play in a band. Who needs, no offense to my former dean, who needs, you know, Shakespearean plays. That's boring. Come see your dean playing a band.
- [01:44]
C
God bless America.
- [01:45]
A
Yeah.
- [01:46]
C
Does this bar feature bingo?
- [01:49]
D
You said it's called Chicken Shit Bingo.
- [01:51]
C
Is. Does it feature bingo?
- [01:53]
D
Can you play bingo at this bar?
- [01:55]
B
You can play bingo. We are. We are waiting for where the chicken will shit. And that determines your bingo number.
- [02:10]
A
Hello, everyone, and welcome back to Rational Security, the show where we invite you to join members of the Lawfare team. As we try to make sense of the week's big national security news. We have a special April 1 episode for you today. No, we're not going to talk about made up topics. It's all for real. But you know, we're going to dig deep into some interesting wonky tech facing topics and I've got a wonderful tech facing panel plus Anna Bauer joining with us. I won't lump you in with that, Anna, but we're happy to have you to talk through some of these things that we are trying to make sense of ourselves and you're going to hear us do it in real time. Joining me for this episode is the aforementioned Anna Bauer from her brick lined beautiful New York brownstone, flyer blazing, just perfect environment with the perfect backdrop. Anna Bauer. Anna, thank you for joining us once again on the podcast and dropping in for this special tech episode and a couple issues outside your lane, but one that's more solidly in your lane. We're excited to have you here with
- [03:02]
C
us for thanks for having me and
- [03:04]
A
join us as well from New York. Filling out the New York contingent, we have Kate Klonik, senior editor here at Lawther. Kate Yu, thank you for coming back on the podcast and digging into some of these tech facing interesting topics we're gonna dig into today.
- [03:15]
C
Always a joy.
- [03:17]
A
And joining us for the first time in a while cause it's been too long is none other than Lawfare senior editor Kevin Frazier from the great state of Texas. He's taken the hat off, he's taken the boots off, he's relaxing with us. Thrilled to have him. Kevin, thank you for joining us today.
- [03:30]
B
Always a pleasure, Scott, thanks for having me. And glad to be back.
- [03:33]
A
Wonderful, wonderful. Well, let's get right into it. Our first topic today, our three topics today as we deal at topic one, the ex post facto rule, the Justice Department and lawyers representing Anthropic faced off last week in Northern California courtroom over whether Defense Secretary Pete Hegseth's ex post and some related communications amounted to an official order and if this Pentagon supply chain risk designation retaliate against the company's First Amendment protective views, among other legal issues. On March 26, Judge Rita Lynn in that case stayed the supply chain risk designation ruling. The Pentagon had in fact unlawfully retaliate against Anthropic. We're also waiting for another related decision from a D.C. circuit panel expected to come down really pretty much any time now. What do we make of Judge Lynn's ruling? Do we expect the D.C. circuit to follow suit and what does it all mean for AI companies and their relationship with the government. Topic two, straight out of Options Oil, gas, helium, pharmaceuticals, fertilizer. The ongoing conflict with Iran has upended global supply chains across the board, with the Strait of Hormuz remaining closed more or less, with some exceptions, as critical infrastructure in neighboring Gulf states faces Iranian attacks. The US has started to feel the first of the effects through rising costs and a trepidatious stock market reminiscent of the supply chain shortages that we felt during the coronavirus pandemic, among other incidents in the recent past. Unclear how severe and how long these shortages will last, but what could be some of the national security and political implications if these shocks continue? And what does it mean for the trajectory of the Iran conflict? And topic three, Space, the financial frontier. NASA astronauts are hoping to launch this week, actually later today on the day. We're recording on the Artemis 2 mission, the first crewed mission to orbit the moon in more than half a century. It is the biggest step to date in the new emerging space race, most specifically with China, one driven predominantly by private actors. The biggest whom SpaceX is preparing to make an unprecedentedly large initial public offering in coming weeks. How should we be able feel about this new, very different space race compared to past ones, and what might it mean both for good and for ill? So for our first topic, we are going to revisit this anthropic case that we have all been following. Anyone in the national security space has been following very closely. I think it is officially rocketed to the top of the charts of major national security actions happening in the courts this year. Anna Bauer, you were at or listen in on, I should say, one of the hearings that we heard in the Northern District of California, really I think the only hearing we've heard on the matter of substance so far. Talk to us a little about what you heard. What gave you a sense of the posture of the parties of the judge in that case that we now have a ruling in which we'll get to in a second. But first, give us a sense of your impressions from that hearing.
- [06:00]
D
Yeah, so Judge Rita Lynn, who is the presiding judge over this case, held a hearing in the Northern District of California. Shout out to the Northern District for allowing reporters and members of the public to listen in on proceedings. We love it. So I was able to listen in on that, you know, ahead of the hearing. One thing I'll note that was interesting, keep in mind, Anthropic has raised a number of different issues in its litigation, so it's raised First Amendment retaliation claims due process claims, claims that this supply chain risk designation was arbitrary and capricious, ultravirus claims. So it wasn't entirely sure where exactly the judge was going to fall in terms of finding which of those arguments are most or least persuasive. And ahead of the hearing, she issued this notice to the parties about a number of questions that she had. And those questions, none of them really indicated that she was focused on these First Amendment or due process claims. And I know that that surprised some of us at lawfare, Kate, especially, I think, to my memory, because it seemed like maybe those First Amendment retaliation or due process claims might be pretty strong ones. And so it was interesting to see that maybe the judge wasn't so focused on those and was focused more on some of the statutory interpretation questions. But then when we got into the hearing, Judge Lynn started with this kind of like, preamble kind of speech, basically, about the case. And it gave you a sense that actually maybe she is focused on these First Amendment retaliation and due process issues because of the content of that speech. And then later, when she issued this order, it looks as though part of that speech she gave during the hearing was kind of maybe a draft of the start of this order, because she says some of the same things about, you know, it's not up to her or two courts to decide who the government is contracting with, but the question in this case is whether the government went further than that. Also at the hearing, I thought one of the more interesting, or I guess I should maybe say amusing parts of the hearing was the questions that she had about Pete Hegseth's tweet, because keep in mind, there are, like, three communications at issue here. There's the Trump presidential directive that he posted. There's the Hegseth tweet in which he says something to the effect of, like, you know, no one's allowed to contract with Anthropic, and if you do, like, we're not going to contract with you. And this decision is ultimate and final. I'm paraphrasing. But that's kind of like the basic thing that he says. And then also at issue is this supply chain risk designation, the actual letter that was sent to Anthropic. And with respect to that Pete Hegseth tweet, which is kind of created basically this secondary boycott effect, There were a lot of questions from the judge about that. And the government basically had to take the position that he didn't really mean it, Judge, when he said, this decision is final and ultimate and binding, and it has no Legal effect. And actually that doesn't really mean anything. And so the judge was like, are you saying that this, a statement by the Secretary of Defense is false?
- [09:32]
A
Not false, just not to be taken seriously.
- [09:35]
D
Right. And DOJ kind of had to, you know, very carefully tiptoe around saying, yeah, that was, this is not true because. And it was just really awkward and, and kind of fascinating to listen to. But, but as I mentioned, you know, overall, by the end of the hearing, we really had a sense that Judge Lynn is going to rule for anthropic on the, on the PI motion. And sure enough, a few days later she posted an order in which she entered a PI And I maybe will leave it to Kevin and Kate, who I'm sure have thoughts about it, to talk about that order in more depth.
- [10:16]
A
Kevin, let me come to you on that. Yeah, I mean, talk to us a little bit about what Judge Lynn decided, kind of where the procedural posture leads this case. We have an administrative of it for a few days while the government pursues an appeal. So I don't think it actually kicks in until tomorrow when we're recording. So we'll record April 1st. It'll be April 2nd. But so talk to us about what the grounds are that we see Judge Lynn ruling in on and your sense of them. I mean somebody who's been looking at this issue set where they're stronger, where they might be weaker.
- [10:40]
B
So I'll shout out the Northern District as well as, as Anna did, because I think that Judge Lynn and perhaps her clerks are working nine nine sixes because they got this out 43 pages in 48 hours, which was remarkable and kudos to them for, for putting in the work on a very important case. So in terms of the procedural posture, just to go to full law profess a second, for folks who aren't aware when we're analyzing something for perhaps injunctive relief, the standard we're going to look to as to whether or not preliminary injunction is going to be granted by a court is whether the plaintiff is going to be able to establish that they're likely to succeed on the merits, that they are likely to suffer irreparable harm in the absence of preliminary relief and that the balance of equities tips in its favor and that an injunction is in the public interest. So those are the four factors that the court is going to look to in the case here. Because the non moving party is the government, we merge those final two factors. So really we're assessing anthropic arguments on three different inquiries and here the judge was looking at three different potential arguments under that preliminary injunction standard. The First Amendment claims, a fifth Amendment claims, and the Administrative Procedure act claims or the APA for all those admin law wonks. And under each of these different prongs, Judge Lynn found that indeed injunctive relief was warranted in this instance. And as you pointed out, Scott, we're seeing that there's been a seven day period for which the government can appeal this determination. We'll find out what happens presumably tomorrow, which too bad we can't have back to back rat sex because that sounds like a incredible way to have a week. But I will say that, and I'm looking forward to getting into the weeds on this with Kate and with Anna. The First Amendment claims here and the First Amendment arguments are very intriguing because we're diving into what exactly Anthropic said and did in relation to this contractual argument and when we cross from conduct into speech and expressive activity. And as much as I'd like to applaud Judge Lynn for doing humans work in a rapid fashion, I will say that of the 43 page order, like three pages are reserved for this First Amendment analysis, which is quite extensive and I think needs more inquiry and needs more deliberation. Again, based off this procedural posture, it's not surprising that it's short, but I think this is going to be a key area of contestation. Similarly, on the Fifth Amendment claims, there's a real interesting scholarly debate and scholarly discussion that needs to be had on this idea of what anthropics liberty interests are. So if you're going to bring a procedural due process claim, you need to show some deprivation of life, liberty or property. Here the judge looked at the sort of ability to offer and make available an AI product as a sort of liberty interest. And this is not something I've necessarily seen advanced in any law review article and I do my best to keep up with what the heck folks are writing. But this sort of notion as an ability to advance and offer an AI product as a sort of liberty interest and make sure that you're able to offer it in a specific format pursuant to specific safety standards, so on and so forth, is a very interesting argument that I think warrants more investigation. And then finally, I think the perhaps easiest case that I'm looking forward to diving into as well is this APA claim where really we need to make sure that when you are seeing some sort of final action being taken by the government, whether that's the Hegseth Declaration or not, the final action here being the supply chain risk designation under two different statutory authorities. But here we're only focused on one of those, whether the explicit procedural processes were followed and whether there was a sort of arbitrary and capricious decision that was made. That seems to be arguably the strongest basis off of which Anthropic can say the government fell short of its obligations under the statutory authority upon which the Department of War, Department of Defense was acting. For example, there needs to be a determination that the government considered lesser alternatives to declaring a party as a supply chain risk in the letters that were sent to congressional committees as mandated by the statute. It was just a conclusory analysis of saying, yes, we considered these lesser alternatives and indeed they weren't available. But that's the sort of analysis that the courts do not look kindly upon. So those are the three different prongs. There's a heck of a lot to dive into in those different domains, and we will wait to see what the government has to say in rebuttal or on its appeal. So stay tuned for tomorrow because I'm sure Lawfare will be pumping out some great insights.
- [15:41]
A
So, Kate, talk to us about your sense of this case. Somebody who's also been watching. We had a little conversation on the side talking about some of the way the judge approached and prioritized some of these arguments. What stood out as you kind of interesting about this case, where the judge came out, what kind of caught you by surprise? Where do you think it might lead?
- [15:58]
C
Yeah, so there's a bunch of stuff that I think is kind of interesting. One is, is that and we've kind of headlined this before, is that when the judge wrote about when Judge Lynn in the Northern District of California posted the questions that she wanted everyone to have answers for and be briefed on ahead of the hearing, there was very clear that they were not First Amendment based. Alan and I kind of had a back and forth about like he was like, I told you that like she's going to skip over the constitutional questions, Constitutional doctrine of constitutional avoidance. We're not going to get into these types of things. I mean, it makes a lot of sense, frankly. But then as Anna kind of like headlined and like kind of summing this up, she did like, that was like the whole thing. She was basically like, I want to talk about all of these things, but this is all kind of in service of the fact that I think that there's a First Amendment issue here. It's very clear. So here's the interesting thing about the First Amendment issue, as far as I can tell, in the briefing that Anthropic prepared and in the. There was an insane amount of, of friends of the court amici briefs filed so quickly in this case. A number of them came from, like, advocacy groups for the First Amendment and things like that. A lot of it was around the speech issues, specifically, what's going to be a hot button issue as we go forward in all of this AI stuff, which is trying to figure out how the law designates code, whether it's going to be seen as conduct or whether it's going to be seen as speech. And that's going to have a huge role in how it's regulated. But anyways, all of that was kind of interesting. But the really interesting thing was the retaliatory claim on the First Amendment actually, if you kind of squinted at it and looked closely, was like a little bit questionable the way in my view, how Anthropic argued it in their briefs, which was kind of like they were saying the retaliation came because Anthropic canceled these, like, backed out of these contracts. And that means that it actually, the retaliation actually wasn't for speech, arguably, that the retaliation was for conduct that they ceased to have these claims. And that is the liberty argument that Kevin kind of talks about and everything else, but, like, it's a harder claim to make on a First Amendment basis. And the judge did a really nice job here of being like, no, the retaliation was in fact, because, like, they were angry. She focused in on a couple of statements that Trump and Hegseth had made on Twitter that was basically like, they were angry that that Anthropic went to the press. And so that was speech. Right? So that was like, it was not that they can't like that they refused to code exactly what that they could. And that code is speech or something like that, which would be like that. Like what the, what Anthropic was kind of arguing or pushing for. It was rather that the statements that they had made to the court following following this cancellation, that was their, their, their right to criticize the government basically, after the canceling of this contract and to speak out again about what had happened to them was in fact the protected speech that the judge found was retaliated against. And I think that this is, this is pretty significant. I think it was. I think it's a pretty strong First Amendment argument, frankly. And I think that, I mean, all of the other, like, arguments are strong. Statutory arguments are very strong here, like, at the very high likelihood of succeeding on the merits in my opinion, I think everyone's opinion I've talked to, but she reached this first in her decision, which is quite long, as Kevin said, for, for something written so quickly. And it went, it went basically First Amendment dpc, like due process clause, Fifth Amendment kind of arguments, and then it went into all of the statutory stuff. And so everything's preserved. It's not really like super significant on appeal, like what they dig into, like they'll, you know, the, the appeals court can certainly reorder like the like or decide, you know, that one that like any of these things are enough to grant a preliminary injunction on. But I just thought it was interesting that that's how she kind of dealt with it and that she thought the First Amendment claim is the strongest and ordered it first. The other thing that I'll say is that we have not yet seen the government file an appeal on the stay that was issued on the preliminary injunction. Judge Lynn, when she issued the preliminary injunction, said, okay, I'll give you a stay for seven days and that will give you time basically to like get your to. To start this appeal. Well, they haven't. And like I'm trying to kind of figure out the strategic posture of this. So I mean, this preserves a stay, preserves everything so that like, that this designation is still in. Like, like everything the government did is still intact. It's going forward. So I guess if they were going to make out some type of claim that they will be irreparably damaged. Like them not getting their stuff together and in time to make this claim is going to be harder to make a claim. Like they've just kind of sat on their hands for six days. I don't know if they're looking for like an extension of this day. I have no idea what's going on. So it's very strange. Meanwhile, and I will also flag this, we're a week out from the end of briefing in the D.C. circuit. And in that case the anthropic flipped the order of their arguments, interestingly, and how they argued them in the brief. And I think that that was in like, I think that that was. I don't know why, but I think that you could just kind of argue that that was strategic in terms of the panel that they pulled, although they had filed their complaint before they pulled their panel. So I think that those are all just kind of interesting things to kind of be looking for. End of the day tomorrow we'll see if the government files an appeal. And I think that you heard me say last time, Scott, That I think that, like, they don't want bad law in this. Wouldn't it be crazy if they don't make appeals? Like, the government doesn't make appeals on this stuff. And everyone is kind of like, no, they're going to go to the map because this is like, like, this is like machismo bullshit, like, kind of like thing. Like they're gonna like, defend this, like, I don't know, like, do you want to, like, make it really clear that you can't use, like this random, like this random designation, like, in the future? Like, I don't know. Like, I don't know. I'd be interested in, like, what Kevin thinks about that.
- [22:07]
B
So it's worth noting that the government based its supply chain designation risk based off of two different statutory authorities. And only one of those was subject to analysis by the Northern District, and the other one is going to be investigated by the D.C. circuit. And Undersecretary Michael has insisted that the supply chain risk designation is still in place under the other statutory basis that hasn't been ruled on in the D.C. circuit. So perhaps we're going to see the government just put all of its eggs in the DC basket, Easter pun intended. And perhaps that's their strategy. I don't know if they just want to give up on the Northern District of California. I think think many people would be surprised if there was a favorable pro government ruling out of the Ninth Circuit, for example. And so maybe this is just trying to leverage its resources and target them in what they may think is a more favorable jurisdiction. But here again, just the facts aren't good for the government. When you go and look specifically at the statutory arguments here, it's worth noting again, what the actual legislative history was of the supply chain risk designation statutes actually is whether or not we'll see the judges lean into legislative history is another conversation. But this is very much focused on foreign actors intending to sabotage the United States. And as Judge Lynn highlighted, it's very weird for a company to, quote, unquote, sabotage the government after months of what the government has labeled, quote, unquote, cordial negotiations. Right? That's not the traditional playbook for sabotaging or undermining the government. And so just when you go through and see, is this the basis for designating a company, a U.S. company, as a supply chain risk, the only AI company that prior to this point was actually on classified networks. It's just a tough road, a hoe. And I'm not sure that on the facts you're going to be able to convince any judge on that. You know, we have clear statutory authority as the government to be able to move forward with this.
- [24:17]
D
Okay, so can I jump in on something on the due process issue that Kevin was, was talking about earlier? I don't think that this is necessarily a completely novel kind of interpretation of the liberty interest factor. The reason that I say that is because if you look in the order, Judge Lynn, she doesn't cite any 9th Circuit case law or really Supreme Court case law, but she does cite some D.C. circuit case law on this point. And I took a quick look at that case that she cites this morning and some of the other case law. And it is the case that the D.C. circuit has developed this what they call reputational harm plus kind of analysis when they're looking at whether someone has been deprived of a liberty interest under the duties of due process analysis. And basically the idea is that the government has inflicted reputational harm on someone or on a company, but that's not enough. You also have to have this kind of broad preclusion to engage in, you know, a calling or in employment. And so what she's doing here is, is applying that established analysis from the D.C. circuit to the case here because she's saying like, you know, not only has anthropic reputation been harmed, but like these, like, for example, the presidential directive, like, quite literally, you know, has a very immediate effect of, of saying like, you know, people can't contract with anthropic. And so I, I get what you're saying, Kevin, that I do agree it's a little bit novel and probably, probably will be the subject of discussion as this case moves forward. But I just wanted to kind of make the case that it's not a completely new kind of thing that she's doing here. She's applying established law from the D.C. circuit, no less. Obviously not binding in the Ninth Circuit, but still, I think significant. And yeah, so I don't know that it's necessarily as novel as you suggested.
- [26:30]
B
Well, so I appreciate you pointing that out and I think you've given me a chance to clarify what I meant to say here, which is, so we see in the order the liberty interest protected by the Fifth Amendment encompasses the right to follow a chosen profession free from unreasonable governmental interference. And then we see the tie in to the idea that depriving a corporation of the ability to receive government contracts, for example, can trigger that deprivation of a liberty interest. And so I agree with you that the analysis that follows that is, is a slam dunk. Applying strong precedent that, you know, although it's only persuasive in the ninth Circuit, does seem very compelling and very logically structured for me. I'm more intrigued in the sort of broader scholarly question of this idea of AI development and AI use as a chosen profession, as a sort of liberty. That's just, to me is. And I should have made that clear more generally. That, to me is a very fascinating and wonderful question because this has been raised, for example, in states like Montana, where they've enacted a quote unquote, right to compute that basically says that the government must survive strict scrutiny before you're regulating computational resources and access to computational tools. And so this whole liberty interest argument and the use of AI and the development of AI is something that I think is novel. And I want to explore further as applied to the question of denial of government contracts. I agree. Alley Oop, slam dunk, LeBron James, all that jazz. I think it's. It's a very strong argument here that's.
- [28:09]
A
That's really useful. Yeah. Because in some ways I was kind of. It's interesting to see her go down this avenue and frame it this way. I see why. Because you're looking for case laws that she didn't find Circuit. But like the idea that the government saying, hey, no one's allowed to contract with this entity that hits core property interests that don't even. You don't need a lot of case law to point out, hey, you're costing business entity. It's not just reputational harm. If it were just like anthropic sucks, then maybe you'd have this question like, well, is simply reputational harm enough to rise to this level? This isn't reputational harm. This is public policy like prohibition.
- [28:39]
B
Just a very short follow up on that. It's worth noting and emphasizing and highlighting and you pick your adjective, just how important it is that Claude has been the only tool used on classified networks for years. And so when you go and say, for example, that Anthropic is operating in an unpatriotic fashion or is trying to sabotage the government, there's just not a whole lot of alignment there. And so also calling forth a reputational harm when you do see the government labeling you as un American and so on and so forth, especially during a time of war. War and geopolitical chaos, I think does have just a clear reputational harm. Now, Scott, as you and I kind of nerded about earlier, this seems to have backfired on the government in terms of interest in Anthropic and A willingness among the American people to shift from other AI tools to Claude, for example.
- [29:34]
A
Well, that's exactly what I want to talk to you about, Kevin. Like talk to us a little about what this has meant for Anthropic and for the biggest market and to some extent DoD, the intelligence community, other entities relying on this. Right. Like you know, as this has all gone on almost perfectly aligned timing wise. We have seen an incredibly sophisticated set of military operations take place in Iran. Highly controversial, highly problematic in my view, in the view of lots of other people, from legal perspective, policy perspective, whatever. But as from a tactical technical perspective, like kind of ludicrous, with the rapidity with which we've seen the United States and Israel engage in targeting something that clearly has some degree of AI involvement given how quickly these target packages are being put together and executed, including some mistakes clearly we have like the girls school tragic bombing that took place in the first day. But frankly that has happened in a lot of wars without AMI involvement. It's more just the scale here of which they're able to accomplish in the first 24 to 48 hours is extraordinary. People looking at this are like, wow, this is like almost approaching a level of warfare we didn't weren't even sure the United States and Israel and other powerful military powers could accomplish. So DOD needs this stuff, is using this stuff. We heard all lot of talk from Trump from Hegseth about a six month implementation period of these bars not reflected in the actual letters and I don't think Judge Lynn really knows what's going on with that either. She mentions that there's no clear description about how this implementation period, if it's happening, is going to be implemented. Do we have a sense of that? And then more generally, what has this meant for anthropic and AI industry companies generally we saw OpenAI pretty early pivot in and both kind of to some extent Sam, by Anthropic's position on principle, but then also go for its market share and say we're going to provide alternative services to DoD. Does that look like it's going to be able to fill the need? Is this affecting Anthropic's ability to compete? Like what are the dynamics around this in the broader AI industry that we see now that we are not quite a month in to this whole travail?
- [31:30]
B
Well, there's a heck of a lot to tackle there and I welcome my other panelists to jump in as they see fit. The first thing I'll say is that Anthropic has said that they stand to lose perhaps hundreds of millions, if not billions of dollars in contracts as a result of this whole affair. And so, just in terms of irreparable harm and injury to Anthropic's bottom line, this does have very severe consequences. And so just want to highlight that first and foremost is why Anthropic is on the defensive here. Secondly, when it comes to the use of AI in a military context, like any good law professor, I'm going to assign some homework to the readers, which is to say you all need to go read Professor Scott Sullivan's piece in Lawfare, which is AI as Abnormal Technology. This is a excellent piece on how in the military setting, the use of AI is just moving so much faster than in other domains. The idea that there's sort of institutional friction or inertia in, let's say, a Fortune 500 company when it comes to adopting and using AI just doesn't exist on the front lines. And we're seeing this play out in real time, as you mentioned, Scott. And so having the latest and greatest and most reliable tools be available to the Department of Defense is a national security imperative. Because these tools must be reliable, they must be verifiable. And Claude and Cowork is doing incredible work making sure that folks have access to those tools and in Iran, in Venezuela, and so on and so forth. And so what folks are debating, at least on X and the other AI blogosphere is the extent to which it's going to be burdensome for the Department of Defense to switch out from Anthropic to other tools. Now there's a traditional sense that, hey, you know, you just swap in a different AI model, bing bang boom, and you carry on and everything works out as intended. As we saw yesterday from the leak of some of Claude Code's code, Anthropic has a very sophisticated approach to making sure these tools work as intended. They have a very thoughtful architecture behind how all of these tools are being implemented. And so whether or not just a sub out into Grox Products or Xai's Products or OpenAI's Products works as intended and has the same degree of reliability and fidelity to user intent is an open question. And so I do think it is important to flag that we're just not sure what a sort of switch to new models may mean for the Department of Defense's functionality, especially at a time of war, that that should raise some eyebrows among the public.
- [34:14]
A
Let us pivot to our second topic, which has a clear intersection here, because we are still talking about supply chains. But now we're talking about perhaps a more credible threat to supply chains. That is the closure of the Strait of Hormuz and the impact it is having on a lot of key inputs into our economy. Kate, you have become one of our self assigned supply chain mavens at Lawfair following this closely. A supply chain nerd. I consider myself a bit of a supply chain nerd increasingly these days as well. Talk to us about what we are seeing and these knock on effects. I think we all know the price at the pump issue which has been the forefront of the kind of media coverage and frankly the day to day experience of most Americans resulting from this conflict. But we're seeing a lot of other knock on supply chain effects to talk to us about what those are and how those are intersecting with the Iran conflict.
- [35:02]
C
Yeah, so I have become like I'm just kind of gotten really into supply chain stuff. I love systems. I love like kind of everything about how they work and figuring out like kind of the entire soup to nuts of them. And so like this is. So as I've been digging into AI and getting into AI, like I have been getting more and more into chip manufacturer and then that also ends up becoming this kind of essential like minerals and gases. Types of like supply chain questions which are fascinating and involve geopolitics and all of these different ways than actual direct regulation on tech does. But I digress to get to your question. There are a couple of things besides oil that are really critical that have been cut off from the Strait of Hormuz kind of being closed. One is liquid natural gas which is has a byproduct that can also which it can of helium which I'll get into which is key in semiconductor manufacture. But then there is also there are also a number of kind of questions about sulfur and polyethylene. A bunch of things that you don't kind of think of as critical. But these kind of ripple on and have these huge effects longer in the supply chain. They affect the development of things like fertilizers, which then affects the cost of food. Obviously these are kind of macroeconomic questions and so I'm not going to kind of get into them. But I've been following specifically kind of just like the helium. And if you don't know about the helium supply chain, it's kind of interesting. Most of our helium comes actually in the world from the US and then about a third of it though comes from Qatar. And so that is the, that, that that one third is the part that's restricted right now through the closure of the straight. And that is I think going to have kind of this huge effect. Semiconductor prices were already going to be like going up. Like there has been like. And memory, global memory shortage is like, you know, which Natalie always reminds me sounds like something very Orwellian. But it's actually about the supply of memory that is available is going to go up, it is going to increase. It was going up before the Iran war and it was going to like it's going to kind of keep going up with everything that's going on. But it's just kind of, it's just a, essentially like people don't want to go to the street. It's highly, it's like either they can't and they're blocked or that it's a, it's a risky maneuver. And so like it's just people are going, not going or going around or there's just a, there's a huge change and a shift. And I kind of think that like I know Kevin's going to speak jump in here too. And I know Kevin also loves these things. We both are also sea cable nerds. I think that we both are kind of drawn to the fact that like all of the sophisticated technology, all of this stuff, and I don't want to speak for you Kevin, but this is my kind of fascination with. It ends up kind of being supported by the most fragile of like architectures and the most human like shit like actual cables that are actually on the floor of the actual ocean. Like, you know, it's like, it's like we can all have all of these fancy things but like at the end of the day so some type of like some type of cable has to get run from like Europe to the United States, like just in that old fashioned way. And so this is kind of the same thing that is so compelling. At the end of the day you just have to ship like helium canisters back and forth to like from, from Wyoming to, to you know, China. And that's like how it has to work. And you have to get it from the ship on the, you know, and then that uses gas and then if you can get the them at all. And all of it is kind of like this crazy kind of thing about the fragility of our entire ecosystem and how much we take for granted. And that's kind of why I'm attracted to kind of this issue and to kind of give the like the listener kind of a view or an argument about why you should Care about the supply chain, I guess, and why. It's actually a really super interesting thing to geek out about, but Kevin, go ahead.
- [38:53]
B
Yeah, no, I mean, I'm a total fellow nerd in this regard to Kate. And I think that when folks realize that modern life wouldn't be possible unless we had all of these seemingly small interactions add up to incredible ecosystems, you do become more attentive to the fact that today my employers aren't going to like this, but I could go to the Oregon coast, take an ax to an undersea cable and the only fine would be $5,000. That's how old our lives.
- [39:24]
A
Sounds like a hell of a weekend, Kevin. Well worth it. I know.
- [39:27]
B
Don't worry, Scott. I've got this whole trip to Bandon Dunes planned. I'm gonna go play 18 and then I'm just gonna hack a undersea cable.
- [39:34]
C
But don't give people ideas.
- [39:36]
D
Kevin.
- [39:37]
C
Yeah, you're like worse than the AI. Like why? Like, what should I. What is the most disruptive thing that I can do to like the global like information ecosystem? And then it's like gonna be like Kevin's face with an axe.
- [39:47]
B
There we go. If this is what I have to do to get Congress to act on undersea cables, you know, challenge accepted. But I want to highlight to the fact that we have to have a more nationwide conversation on this whole AI infrastructure. Shout out to Jensen Huang and shout out to JPMorgan Chase. JP Morgan Chase came up with a seven layer cake for the AI tech stack. Jensen Huang has condensed it to a five layer cake of the AI tech stack.
- [40:18]
A
You're not going to go with a layer dip. And as a text of AI, yeah,
- [40:23]
B
it's, it's getting really delicious really quickly. Folks need to realize that all of this AI infrastructure doesn't happen by accident. And so this crisis is really highlighting the fact that, for example, data centers now are a target in the Middle East. We saw that some of the first attacks were on data centers. And so for the states and municipalities that have questions about data center access, and I just want to say data centers don't only exist for AI. We need data centers for hospitals to run, for commerce to occur, and for selfies to get sent on your phones. If we don't build those data centers here, then pushing them abroad does raise some significant supply chain risk concerns because we are seeing this infrastructure, these forgotten pieces of the AI tech stack be targeted. And so this is a real reminder that when it comes to developing that infrastructure, we have to regard it as critical infrastructure. And we need to be far more attentive to the fact that, as Kate noted, the fragility of all of these things is so immense, even a couple weird actors doing weird things can have massive consequences. Go talk to the folks, for example, at Unilever, which now has a hiring freeze as a result of the supply chain concerns they're seeing with respect to their oil products, soap, shampoo, all of that needs a lot of the stuff that's going on in the Middle east. Also. Go talk to the folks who rely on. There was one article, I think it was in cnbc, on how the pharmaceutical industry is being disrupted by the conflict. They said, don't worry, you know, it's only the folks who rely on, quote, unquote, orphan drugs who should be concerned. And I'm guessing there are a lot of Americans who are thinking, huh, thanks for designation as being an orphan. But yes, I actually do need that medication. Can we please take this seriously? Thank you. So just being aware to the fact that any sort of disruption can have an impact on millions of lives in very sensitive, sensitive ways. I think this should be a alarm for Congress to take these sort of boring issues like undersea cables far more seriously.
- [42:36]
D
Can I just say, like an orphan drug, Just to clarify for people who maybe don't know is a drug that it's like people with rare diseases typically who are, you know, need an orphan drug. So just to be clear that when people say, oh, it's just, it's just the orphan drugs, it's like the people who need it, the medication the most. So, yeah, that's, that's nuts.
- [43:02]
B
Yeah, they'll figure it out though, Anna. You know, they'll get by. There's, you know, they'll just kind of just get a drone or something to go pick it up somewhere.
- [43:10]
A
Well, at the point I would, I would drive home for this, that I think it hasn't gotten picked up on yet. But I think we're going to hear about it a lot more in the week, particularly if this we've heard the Trump administration suggest we're in a pseudo negotiation. Slight easing of tensions with Iran while these negotiations are ongoing. Trump has said, oh, after the end of this period, which he's now extended twice, that's when we're going to hit Iranian energy infrastructure. A hitting Iranian energy infrastructure is counterproductive, certainly on the energy front and the global gas prices front, because that dropping production like we saw happen when Israel hit the South Pars field last week has a big impact on global availability and Supply chains, but also the big question. I interviewed two former intelligence community veterans on Friday for the Lawfare podcast about their view of this conflict. It was a great conversation. I highly recommended Chip Usher and Aaron Faust from the CIA and from the State Department. What one of them, I think it was, Chip said the big question for him that he didn't understand is so far as why the Houthis in Yemen, who are themselves partners of Iran, not quite full clients, but closely allied with Iran today, hadn't really, when we recorded, gotten engaged in this conflict. And they control the other key supply route out of the Gulf area, specifically the Strait of Mandab, which goes up into the Red Sea, so through the Suez Canal, which is a hub of global shipping. This is why during the Biden administration and early in the Trump administration to a lesser extent, that when the Houthis were hitting maritime traffic over the Gaza conflict, it was a huge supply chain problem then. And you think back to like, you know, the, the, was it the Evergreen, the ship that blocked the Suez Canal, how bad a complete shutdown of that can be? The Houthis enter the conflict over the weekend, right? Like at a small scale so far, and we'll see what the level of commitment is. But they started hitting maritime traffic in the Strait of Manda. So this could get a whole lot worse, a lot faster across a whole broader range of goods. Because while the Persian Gulf is mostly about getting in and out things that are exported through this key part of the world, it is a, it's a cul de sac. It doesn't really go any other places. The Red Sea, Strait of Manda, Suez Canal, that's like the key route through the Mediterranean, right. To get to other avenues in other parts of the world, it could have a much broader effect. So we're going to see these hidden blob, other industries moving forward. Before we move on this topic, Kevin, I want to drill in on one thing with you. And Kate, you may have views on this, too. Amanda, you might have views on it, too. I'll invite you in. But I think you may be a little more in dark on this than I am, because it's a more technical AI question, like, what does it mean if you have these helium shortages preventing the additional construction of these chips? What aspects of the AI industry does it affect? Is it the training and development of new models? Is it the broad availability of AI services? Is it the expansion of AI services to other industries? Which I actually think probably blends in with the second question, because not like there are they're giving chipsets away. They're like kind of just expanding services. What is that likely to impact in terms of the trajectory of available AI services, both kind of vertically and horizontally across the market?
- [46:09]
B
It's everything. Chips. Chips are everything. It is like flour going into that cake. If you don't have flour, you're going to get a pretty gnarly cake. And if you don't have chips, you're going to get shitty AI. And so this has to do with not only the Frontier Labs developing the latest and greatest models, but also what we refer to as inference. So this is just your ability to go and query your favorite model for whatever thing you're looking up or whatever task you're trying to accomplish and what
- [46:36]
A
inference means in this case, just now. Yeah, yes, yes.
- [46:39]
B
This is going to become even more significant as we become more reliant on, for example, AI agents. So right now, any individual's need for compute is relatively low. Right. If you're doing fairly unsophisticated tasks like using ChatGPT to come up with your March Madness bracket, not a huge computational load. If you are instead saying, hey, I want Claude to now design my syllabus for the entire semester based off of all of the lectures I gave last term and to integrate the latest news into those lectures, the week of so on and so forth. Now you're seeing the AI agent trying to go through and use those computational resources for hours, if not days. If you go talk to the folks who are on the bleeding edge of using AI, their computational needs are even higher. And so this has massive consequences not only for the leading labs, but I also want to call attention to the fact that we need to support a competitive AI ecosystem. I think it's also important to flag that access to computational resources is incredibly important, not only for the Frontier Labs labs, but also for our researchers, for our academics, for the folks who are doing new scientific discoveries. All of them need to have access to these chips. And we're usually trying to plan months, if not years ahead for the sort of demand we're going to see from the industry. And so even small lapses in the supply of these chips can have huge ripple effects.
- [48:07]
A
It's wild.
- [48:08]
C
You know, I would say that the flower analogy is right. I don't know. There is. It just, just, it just blocked. To put it even, like, in more simple terms, it just blocks the, the, the, the rate at which things can scale. It's just taking an essential element. You don't so like to, to put it another way if you don't have the flour, you just stop making cake. Like you don't like, right. It's not that you make bad cake, it's like you just stop making it. Like you can't make as much cake. And so like right now we are just investing tons of money to make as much AI as possible to make the AI better, to make it faster, to make it deliver things quicker and we'll just do less of that. Like there's just like we won't build the data centers, there won't be the smaller models coming up, there won't be, the bigger models will get more expensive. Everything will kind of. So I think that like that's just kind of like it's a, it's a pretty simple, it's a pretty simple kind of way to kind of think about it which is, yeah, it affects everything. And I, you know the New York Times had a great graphic in this about just kind of all of the downs, downstream effects of like the supply chain. The supply chain all over the world from like people are making DOSA in India because it uses gas ovens to you know, to helium at your kids birthday party is going to be more expensive for balloons. Like this is just kind of, this is, you know, this is the world we live in. Everything is interdependent and people will pass on prices to consumers and this is just very much going to be there. New future.
- [49:30]
A
Well, let us transition to our third topic because some people have some very creative ideas for some of these supply chain issues, particularly around data farms and data centers, including putting them in outer space. And that is one Elon Musk. We may be seeing the first step in this direction to some degree later today at 6:24pm I think is the specific time I can't remember that's east coast or local. The Artemis 2 mission is supposed to launch. If you are a space nerd, and I consider myself proudly a space nerd, this is kind of a big deal. This is the first manned mission around the moon. They're not landing on the moon, they're kind of orbiting the moon. The first time human mission has done that in over half a century. Which is kind of an amazing thing that I don't think most people are fully aware of how much we just have not really bothered to go back to the moon again since the 1970s or early 1980s. And it's pretty monumental. It's also a huge, huge step in a much broader multi year plan we see and have been laid out through multiple administrations. Now for the United States re entering space in a big way and actually treating space as a kind of a growth domain from a strategic, from an economic, from a variety of perspectives. An area where the United States and Americans as well as people from other countries around the world are going to be operating in our lifetimes in substantial numbers. Kate, you said you were a space nerd, so I'm going to come to you first on this. Talk to me a little about how significant you find this. I think you mentioned that you watch a lot of these launches. I tend to watch a few of them too. I haven't tuned in live to this many. I may try to tonight. My kids cooperate to see what exactly happens. But I would just be kind of curious. How significant do you think this step and the broader effort is in the broader picture? In an era where we are seeing such rapid change on so many dimensions, technologically, frankly, politically, so socially too, to some extent, where does that fit into that broader picture?
- [51:18]
C
Yeah, I think it's political. I mean so much of space is political, actually more than you'd think. First of all, thanks for outing me as a secret SpaceX launch watcher, Scott. This is like going to get me canceled.
- [51:30]
A
They're really cool.
- [51:31]
C
On at least like platforms they're cool. I mean I think of the blue sky mobs are already coming after me. I can feel it, basically. I do. I think the privatization of space is, is like, is actually just like absolutely fascinating. And the idea, the privatization of the space industry, what's separate from that is that SpaceX is obviously going to like has an upcoming like, like public offering that is going to be hugely significant. Which is the other piece of this that I love that is super interesting. And the other piece that does tie the space race into kind of my work is the fact that XI was merged while both are still private companies and they don't have to like face any type of, like any type of government scrutiny into the same company, which is a huge thing for the valuation of the platform X and grok, which are subsidiaries of xai. So this is kind of all tied together the space race and everything that's happening with it right now. I think that the significance of this is essentially that like that NASA and a public entity agreement, government entity is putting efforts into space again when it was for quite a long time before we saw billionaires decide to like dip their, to throw their hat in the ring was kind of like this, like why do we need to go to, why are we wasting money on going to the Moon and shooting rockets into the sky when we can, you know, when we have all of these problems back here on Earth. Which is absolutely true, frankly. Like, there's a lot of good reason to kind of of like think that that's something. But there has been like just a tremendous, tremendous amount of like, of advancement and growth with private money once this has been taken out of the realm of NASA. And it is, it is, it is incredible, frankly, to kind of see what is capable. It just kind of boggles the mind. It is the stuff of like, to watch these launches, to watch the chopsticks, to watch everything happen. It's just like kind of these like, just marvels of engineering and like, things that I never thought that I would kind of see in my lifetime. And then I just want to just point out that, like, you know, we also just still have sea cables. Like. Well, like. And so, like, it's just like you watch this gigantic 100 million ton, like kind of spaceship go off, get launched, come back to Earth, Earth get caught on these chopsticks, set down, and then so it can get reused. Just this.
- [54:00]
A
And then you see Kevin in the background hacking an undersea cable with a hack.
- [54:03]
B
Exactly.
- [54:04]
C
And then there's Kevin.
- [54:05]
A
$5,000.
- [54:06]
B
Give me a good axe. Like a good, you know, old Portland Timbers fan. Give me an axe.
- [54:11]
C
What a. What a time to be alive, people. What a time to be alive. It's just the high and low of it is like, just astounding. So.
- [54:20]
B
Well, and this is, this is the book that me and Kate need to write at some point on just the scale of all of these projects almost inherently have to be done from a private sector standpoint. Now, just because we're not seeing the level of funding, I mean, for hundreds of billions of dollars to go towards any of these critical infrastructures now that just doesn't, poof, come up out of thin air. And so we're seeing private dominance of the undersea cable system, we're seeing private dominance of Internet via satellites. We're seeing private dominance of space generally. What does that mean for the rule of law? What does that mean for accountability? What does that mean for international law? These are huge questions. But I do love that for one day we can all nerd out of watching this potential launch and being distracted from a lot of other things, for better or for worse. And just seeing some great Americans. And one Canadian, I will note, there is a Canadian, the first Canadian that will go around the moon. So congrats to our friends to the north for, you know, Getting, getting a maple leaf going around the moon. Exciting stuff.
- [55:26]
A
Well, I will say, because I looked this up before out of sheer curiosity, if you doubt the American government's ability to spend money on big ticket items. Still, I will note the cost of the Iran war as of this month actually exceeds the annual budget of NASA. Around the $25 billion mark equals to exceeds, roughly depending on how you count it. Exactly. So we can still spend money on big ticket items, but not always the big ticket items you may want. Exactly. Only certain select ones. I want to put this in a little bit of context for people about where this fits into NASA's plans as I understand it. Oh, feel free to correct me if I'm wrong on this. So Artemis's ultimate goal is to build a US permanent presence on the moon or in the orbit. That's actually changed recently. There were plans to build a orbital moon base or orbiting around the moon. Initially that's now been phased out in favor of a sooner permanent moon like surface base. On the logic that once you establish that you have a landing platform that because of the lower gravity of the moon and moon orbit, potentially access to resources there, including helium 3 and a bunch of other stuff that you might be able to get out water, you can. All of a sudden the cost of sending things further out in outer space gets way cheaper. Cheaper because you don't need to escape Earth's gravity to the same extent. It becomes a platform for being able to do all sorts of further outer space exploration. And this was some of the debate between the Mars versus Moon race that we saw Elon come in. Elon was basically saying, no, we should go straight to Mars. Forget all the skipping the moon step. And that appears to have lost out now to the more conventional NASA moon base saying this is the next step towards the rest of space, is establishing a platform for which we can launch more cheaply and effectively. Effectively and safely. To some extent, reentry, exiting and entering Earth's atmosphere is still the most dangerous part of any space travel at this point. And the most expensive, most dangerous, somewhat relatively, to the relatively unambitious missions we're taking so far. That would change if you're trying to go a lot further out in outer space. So I guess what I'm curious for you all is like, what, what do you see as the, like, most promising elements of this sort of mission? I mean, in theory, Artemis. Right now, I think Artemis 4, which is scheduled for 2028, is in theory supposed to put human people back on the moon. Right. They're going to have one more mission next year where I think it's next year, where they're going to be testing basically like landing crafts that are being built by Blue Origin and by other folks, but where theories have people back on the moon in 2028 and then move towards this more permanent habitation of Syndrome expansion. Is this an economic drive? Is this commercial drive? And what are the risk factors that you all see? I mean, what makes you anxious about it? What makes you excited about this possibility, about seeing this, us again in our natural lives, which is kind of. Kind of extraordinary.
- [58:02]
B
Yeah, I'll say that. I think we have to expand our horizons for, for lack of a better phrase, in terms of finding new domains to advance scientific knowledge, in terms of finding new new domains for these infrastructure endeavors. Whether or not I would gamble my own very limited funds on the creation of data centers in the next decade in space or a manned moon base in the next decade, I'm unsure. I do think, though, that we are living on a dying planet in many regards, and finding the sort of resources and advances we need to be able to sustain all of the things that we hold dear does require scientific advances and does require new and novel insights. And so in that regard, I'm very optimistic of trying to push us to new frontiers and trying to expand our capacity, capacity to study. Study space and see all that we can leverage there. I will say, though, that this is going to become a very fraught area very quickly, to the extent it isn't already. All Law of the Sea nerds love to say that they understand a little bit of space law, because that tends to be the law of the sea is the law of space. And for all those who know anything about the law of the Sea, we know it's not going to particularly well, especially with respect to deep sea mining. That's becoming more and more prevalent. And so as soon as we do find out there's a vast repository of X, Y or Z rare earth mineral, or I guess, do we say rare space mineral? I'm not sure what the. The right phrase would be. It's going to get contentious and gnarly and legally ambiguous very quickly. And that doesn't add up to chill geopolitical vibes. And so that, that does give me pause because we're already seeing that countries, for example, even India, are capable of identifying and targeting things in space and blowing them up intentionally, therefore sending tons of debris. That becomes a huge barrier to more space exploration and more space endeavors. So even this question of space debris hasn't been figured out in a meaningful capacity. And that can be a real cap on our potential to make use of space. So I'm at once very excited that we are pushing the horizon and that we have folks who are asking big questions. But I worry about the nitty gritty questions of just is there even an orbit for some of these more sophisticated and societally important areas for, for us to explore?
- [60:31]
C
So I agree with that. And I will also just add the one other, one other thing that I think that is really true of everything happening in space is that it's quickly becoming a monopoly or already is a monopoly. Musk owns two thirds of the satellites in orbit right now, and so through Starlink or other subsidiaries of Starlink. And so there, I mean, that is something that is not great as we increasingly shift away from undersea cables and to satellite provision of Internet services or telephone services. So I just think that having kind of our eye on what's happening in space is going to be increasingly important.
- [61:10]
A
Well, with that, we are out of time for today's conversation ran a little short on that third topic, but we have many opportunities to revisit it as I'm going to be watching in the weeks to come. But of course, this would not be rational security if we did not leave you with some object lessons to ponder over in the week to come. Kate, what did you bring for us this week?
- [61:27]
C
Well, it's topical, what we were just discussing and have discussed throughout. Like the man, like just I wish I could stop talking about him. But my friend mailed me this the other day that she got. And this is my object lesson. It is a notebook. I wish it was like a book. I mean, it could be a book. It could be a book many, many pages long. But like those who can't see, it's a book that says things Elon Musk has done to piss me off. And it is a notebook. It only has a certain number of pages. So I, you know, but I have not written anything in it yet. But yeah, I think that that's my object lesson for the day. There's a constant kind of question for me is like, how we deal with the new billionaire power structure that is fueling all of tech and increasingly all of geopolitics and national security. So that is, you know, that is unfortunately my object lesson for the day. To leave it on an optimistic note.
- [62:22]
A
Well, I think we have lots of opportunities to make use of that particular object lesson in the weeks, months and years to come. Anna Bauer, what did you bring for us this week.
- [62:31]
D
My object lesson this week is Survivor season 50. I know that the reality TV show Survivor is very boomer coded, and people might be thinking, are there still people who watch Survivor? But we do, in fact exist. And every week I actually go to a bar in Brooklyn that hosts a Survivor watch party, and there's like, almost always like 100, you know, 50 to 100 people there who show up, and it's super fun. But I recommend this season in particular. I first got into Survivor during the Pandemic, and my friends and I played this, like, game of virtual survivor. But season 50 has been incredible thus far. It is a season that involves, like, a bunch of returnees from previous seasons.
- [63:26]
C
So if.
- [63:27]
D
If you want to get into Survivor, this is a good season because it's a really good one. And so I hope people check it out.
- [63:36]
A
Well, that's a great boomer recommendation from Anna Bauer. Coming in 50 seasons. I must. They're obviously making more. It has not been on TV for 50 years. Right? This is just because it's not like the last time we were on the moon is when they started doing Survivor. They must squeeze a couple in per year for some of these years.
- [63:52]
D
No, it's definitely because they have multiple per year. I would be really upset if it had been 50 years since Survivor first started, because I started watching it when I was like, I don't know, probably like 11 or something. It was a. Definitely a 2000s.
- [64:11]
A
That's what I thought. Okay. Yeah, I think it might have been around in the 90s, actually, because I feel like I remember watching when I was little.
- [64:16]
D
We are not that old, Scott. That is not that old yet.
- [64:19]
A
It's coming.
- [64:20]
D
But also I want to. I want to tell you, though, for people who are wondering it, it does have a renewed viewership in, like, Millennials and Gen Z. There's not a boomer in sight at this bar that I watch Survivor at on Wednesday nights. So just for anyone thinking that it
- [64:42]
A
is Boomer, that may be more of a bar in Brooklyn in the evening. Phenomenon, I will say, but I'm not a boomer, but I haven't been to a bar in the evening or the week forever.
- [64:51]
D
I'm telling you, everyone is there to watch Survivor. I promise.
- [64:55]
A
All right, all right, all right. Well, for my object lesson this week, I will dip into Television Zone as well for my pseudo annual. Maybe like every other year, depending on whenever the new season comes out. But it's very well timed with the Artemis launch this week. The new season, what is probably My favorite TV show I think is back this week. That is For All Mankind. The first episode actually I think dropped last week. I have not gotten to watch that first episode yet. I like to wait for them to pile up because I hate having cliffhangers. So sometimes I will wait for the whole season to come out so I could just binge it. Especially because I don't really get to watch TV in my day to day life. So I have to wait for like a vacation or sometime when I can just kind of like power through all of them. But we'll see if I can hold off this time long. It's a phenomenal show. I will say. I think the end of season two for all Mankind. If you watch it, you have to watch it. It's amazing. Give it a few episodes. The first few episodes are a little slow of season one. Then it gets you. And the conclusion to season two, the last two episodes, particularly the last episode, which are both film length, they're like 90 minutes. May be the best television I have ever seen. The most edge of your seat intense thing I've ever seen on tv. And it's absolutely amazing. The last few seasons have gotten a little weirder, but I still really enjoy it and I feel like I'm excited to see what the season has in the Pike. Not to mention this weird spin off they're doing about Star City, which I don't know how to feel about. But at least this season worth checking out with that. Kevin, bring us home. What do you have for your object lesson this week?
- [66:08]
B
Yeah, well, I'll double click on For All Mankind. Great show. And my wife and I debate over the do you wait to binge or watch episode by episode? I am an impatient individual, so I'm just like, if it's there, I need to consume it. Um, so yes, the Internet's working out for me wonderfully. I want to bring as my own object lesson. Speaking of mental health and issues, wada, I'm trying to make Fetch happen, which is we've heard of AI as normal technology. Scott Sullivan said AI as abnormal technology. I'm going to call attention to AI as boring technology. And these are all the use cases of AI that are just mundane, that are boring, that no one is really screaming about or clamoring about. There's this new app called Flourish that was sent out to 486 undergraduate students at Harvard. And over several weeks, participants in the treatment condition reported significantly greater positive affect, resilience and social well being. So I am a big stand for mental health and mental health awareness and the fact that we have AI tools that have demonstrated positive use cases is something that I just want to scream from the rooftops because not everyone has access to a therapist. And we should celebrate and welcome the fact that there are innovators working on tools to help folks who have mental health issues. And so kudos to flourish, kudos to HBS students, and kudos to Harvard on this project.
- [67:36]
A
Oh, great recommendation there with that. That brings us to the end of this week's episode. Rational Security is of course a production of Lawfare, so be sure to visit lawfairmedia.org for our show page, for links to past episodes, for our written work and the written work of other Lawfare contributors, and for information on Lawfare's other phenomenal podcast series. While you're at it, be sure to follow Lawfare on social media wherever you socialize your media. Be sure to leave a rating or review wherever you might be listening, and also sign up to become a material supporter of Lawfare on Patreon for an ad free version of this podcast, among other special benefits. For more information, visit lawfaremedia.org support our audio engineer and producer this week is no osband of Goat Rodeo and her music, as always, was performed by Sophia Yan and we were once again edited by the wonderful Jen Patcha. On behalf of my guests Kate, Kevin and Anna, I am Scott R. Anderson. We will talk to you next week. Till then, goodbye.
- [68:23]
C
It.