Pod Insights Sharp Tech with Ben Thompson (Preview) The Anthropic Saga Continues, Fox and the Future of Streaming, Q&A on ChatGPT, Agentic Shopping, Autonomous Driving About this episode On today’s show Ben and Andrew begin with a hydration break take as the World Cup continues before turning to the latest on Anthropic’s export control saga, what the Trump administration may not understand about AI cybersecurity, and what Anthropic still doesn’t understand about dealing with the government. From there: Fables differentiation, the dangers of ringfencing frontier AI within U. S. borders, and how open source Chinese AI may affect the marketplace. From there: Why Ben gives Fox the benefit of the doubt after its Roku purchase this week, whether Siri AI will nuke ChatGPT’s market, Noam Shazeer joins OpenAI, what e-commerce will and won’t be automated, and whether autonomous driving can scale on ICE vehicles.
Transcript Speaker names marked ? are inferred from speaking time and the show’s known hosts — they are not voice-verified.
0:00 Andrew Sharp? Hello and welcome to a free preview of Sharp Tech.
0:08 Andrew Sharp? Hello and welcome back to another episode of Sharp Tech. I'm Andrew Sharp and on the other line, Ben Thompson. Ben, how you doing?
0:19 Ben Thompson? Sad, Andrew.
0:20 Andrew Sharp? Why?
0:21 Ben Thompson? Sad because been forced over the last week to use these really crappy AI models.
0:30 Ben Thompson? Only 5.5 in Opus. Where's Fable? I told you Fable was good.
0:35 Andrew Sharp? Somebody needs to jailbreak Fable back to life for us, man.
0:38 Ben Thompson? We did need the 48 hours of truly leap forward in terms of model capabilities.
0:44 Andrew Sharp? Poor Ben, his wings have been clipped here. Yeah, now here we are, living in the Stone Age. But, I mean, look, only a matter of days until the rule is reversed and Fable is brought back to life would be my guess.
0:59 Ben Thompson? That's what I was gonna say. Is that Andrew Sharp reports or?
1:03 Andrew Sharp? No, look, just as I, it occurred to me as we come on to record to talk about the export controls and everything else. Like by Friday night, I'm sure there will be a twist that renders half of this obsolete. We're not gonna start with Anthropic though, actually, because you informed me that you have a World Cup take related to the hydration breaks.
1:24 Ben Thompson? Oh, you wanna, no, I don't know if we could do this at the top of the show. Because, no, I'm concerned I'm gonna make people so mad that.
1:34 Andrew Sharp? They will hit stop and unsubscribe.
1:37 Ben Thompson? Right.
1:38 Andrew Sharp? Okay.
1:38 Ben Thompson? If it was at the end of the show, then they might go and unsubscribe, but then at least they've listened to the whole show. But if we put it up front, they might not listen to the rest of the episode. Although maybe if we make the rest of the episode really good, then they'll decide to continue to be subscribers.
1:51 Andrew Sharp? A bit of grace after 90 minutes. Fine, we can save it to the end.
1:55 Ben Thompson? I'll drop it. Okay, here you go. Here's my World Cup, uh, World Cup hydration take. So you remember previously, and people sort of got annoyed at this, but I still stand by it. The US implements these policies in the stupidest way possible, right?
2:10 Ben Thompson? So like I said, like we have universal healthcare. It's that everyone has to be served by an emergency room. That made a lot of people mad, but like I stand by it. Like the fact of the matter is everybody, if you, if you're in an emergency, you're going to get healthcare. Like that is universal healthcare in the stupidest way possible. Oh, it doesn't cover preventive. Yes, I know. It's dumb.
2:30 Andrew Sharp? In a very narrowly defined way, it's universal healthcare.
2:33 Ben Thompson? The point of making that analogy is that it's a really stupid way to do a government policy. Or if that was in the context of AI data centers, they're going to just end up paying people to build them. And that's how we're going to backdoor into AI, like putting money in people's pockets, right?
2:51 Andrew Sharp? Right, yeah.
2:51 Ben Thompson? Yes. So this is the stupidest way possible. So this is sort of context for my hydration break take. Which is soccer is, and it is called soccer. By the way, do you know where the word soccer comes from?
3:07 Andrew Sharp? I have no idea where.
3:08 Ben Thompson? It comes from England, to be clear.
3:11 Andrew Sharp? Great.
3:11 Ben Thompson? So, right, football was a sport played on your feet as opposed to on horses. And then there was association football. So C C was soccer, came from England. They, you know, I'm sorry that we preserved English culture better than they did and use the word soccer.
3:29 Andrew Sharp? And they've been getting it wrong for 150 years. What the hell happened?
3:33 Ben Thompson? That's right. Okay, no, that, wait, that's not the take. Don't, don't unsubscribe yet. Give me a couple more minutes and I'll really make you mad. So soccer has all these traditions, the way it's always been played. And, you know, and one of them is you have 45 minute halves, the clock doesn't stop. And that's just like part of the, and the whole thing with the hydration break is, well, the US is really hot. Number one, come on, it's not that hot. Seriously. This is tied into the like, look, go to Taiwan in the summer. That's hot. I would come back to Wisconsin every year. How do you go to Wisconsin this summer? It's so humid. I'm like, it's not humid at all. But it's not that hot, but whatever. Everyone's like, it's so hot. We need to have, you know, some games. It's going to be too hot.
4:15 Ben Thompson? Players need to stop for their health and safety and get water. Oh, if we do it in some games, it won't be a very good for all the games. And actually, what it's actually doing is there's an extra commercial break in the middle of the game. And it's like a commercial break or you're not going to change the channel because you're like in the middle of the game. It's a huge moneymaker. Everyone's upset about it. You're destroying soccer just to make a few bucks. Blah, blah, blah, blah, blah. Um, here's my take. We're actually fixing soccer.
4:43 Andrew Sharp? Okay.
4:44 Ben Thompson? The like, so one of the complaints about the hydration break is the momentum changes. Like a team's like on their heels and they come back out.
4:53 Andrew Sharp? Hate it because it's actually changing some of the outcomes of these games. Because guys come out of it, like I think Holland scored right after a hydration break. That's Erling Holland, the player.
5:05 Ben Thompson? The England Croatia game yesterday, Croatia was like, like on their heels. They come out, reset, score a few goals and guess what? England was definitely the better team in Croatia. England was incredibly impressive in that game. That game was so much more entertaining because of the hydration break, allowing Croatia to reset, score a couple of goals, made it very gripping. England ended up winning four to two. It made the game better. Guess what? Quarters are a good idea. Hydration breaks, a blatant sort of like grow up. It's not that hot out. Combined with a pure money grab. We're backing into making soccer better and everyone should thank us for it.
5:44 Andrew Sharp? Fewer exhausted players is a good thing is I think the kernel of the take.
5:50 Ben Thompson? I like to see people performing at their best. Uh, sue me. So, uh, they're just anyhow, um, don't unsubscribe, but that's my take.
5:58 Andrew Sharp? Sue Ben, but look, my take on hydration breaks is that they should just be called commercial breaks. So maybe we'll get there by the time we get to the next World Cup. Because I do understand the instinctive rage at branding this profit optimization strategy as a player safety initiative.
6:16 Ben Thompson? There's a lot to be said for just being honest. Okay. And guess what? You know what's good for the game? If everyone's making more money, like, like you get more money, like you think about soccer.
6:27 Andrew Sharp? We're all adults here. Everybody else has commercial breaks. It's fine. Um, and I hadn't considered the idea that actually it's going to make the second halves more entertaining than they would be otherwise. So as far as I'm concerned, not a bad hydration break take, uh, soccer fans of the world or football fans of the world.
6:45 Ben Thompson? If you're like super excited for the game, imagine you have the US is playing tomorrow, Australia, maybe you've already watched it by the time you listen to this. Maybe you're pre-partying, you're getting ready. You know what you might need? A dehydration break. And so you're not so bad in the middle of the half, run to the bathroom.
7:02 Andrew Sharp? There we go.
7:03 Ben Thompson? From my perspective.
7:04 Andrew Sharp? Well, my number one World Cup take, uh, you created a giant World Cup group chat and there's like 15 different people in there. You added me without asking me. So I've been in there
7:16 Ben Thompson? Cause you were dropping World Cup, you were, you were dropping World Cup takes in another group chat.
7:20 Andrew Sharp? I was dropping Champions League final takes a couple of weeks ago, not World Cup takes. But being in the, in the group chat, I'm fairly busy. I've got two toddlers, a bunch of podcasts. You are busier than I am though, and I am flabbergasted by how many World Cup games you have consumed over the past week or so. You continue to be the biggest sports fan I've ever met. So congratulations.
7:45 Ben Thompson? Well, thank you. I love the World Cup. It's great. Games are on all day. The, uh, but you see, uh, the Hu Shijin, the, the, uh, the Chinese propaganda.
7:52 Andrew Sharp? No.
7:53 Ben Thompson? Oh yeah. He put out a tweet complaining about the game times that they should be catering to China as far as when the games are played. The games are being played from like 11 in the morning till like one in the morning.
8:05 Andrew Sharp? Right. Relax. 9pm Pacific. Some of the games are clearly being catered to China. So China's being taken care of here and you're watching almost every single one. I'm getting texts from you throughout the day about different World Cup games. I didn't even know were happening. So, uh, I look forward to joining you probably by the time we get to the knockout round.
8:24 Ben Thompson? What's great about the World Cup, I'm not actually a soccer watcher. I always like every World Cup, like, oh, I should watch more soccer. Now it doesn't happen. Like the whole week, I'm sure it's great. I appreciate the quality of play with teams that play together all year.
8:37 Andrew Sharp? Whatever.
8:37 Ben Thompson? I get it.
8:38 Andrew Sharp? It's where you're at. Okay.
8:39 Ben Thompson? No, I am all in on the eventization of sports. I love it. It's like tennis. I watch the big four. I don't watch your regular tennis match. Yeah, you know, I will watch like the big golf tournaments. I'm not watching it every week. I am all in on, I want events. I'm in on the events. I enjoy the atmosphere around the events, the social media on events. And I don't want to deal with it the rest of the year. So it's perfect.
9:04 Andrew Sharp? There you go. Well, I would say I'm halfway in on the World Cup, but all the way in on America. And I had a great time watching USA's opening match and will be watching on Friday night in Seattle. But for now, Ben, we have news to get to. We did talk Anthropic and Fable last week. But as I'm sure 98% of the audience knows by now, late Friday afternoon, the Commerce Department issued an export control directive to suspend all access to Fable 5 and Mythos or Mythos by any foreign national, whether inside or outside the United States. So Fable was-
9:41 Ben Thompson? Or inside or outside of Anthropic.
9:43 Andrew Sharp? Indeed. That was part of the problem as well. Fable was pulled from the market altogether, remains off the market, as you noted at the top. And as we record this Thursday afternoon, that directive is in place. We'll see how much longer it lasts. What do you think of where we are right now? It's kind of a frustrating story to talk and write about because on one hand, it's the biggest story in technology. It also seems like there's a good bit of non-public information that's informing behavior here. So how do you wrap your arms around where we are?
10:17 Ben Thompson? It's hard to wrap your arms around it, in part because what you said, they're like, what do we not know? It's also hard because everyone in this situation is incredibly frustrating to me on a personal level.
10:31 Andrew Sharp? Great.
10:31 Ben Thompson? So, because I, so I actually feel like, I almost feel this week, I took the opportunity to just write about Anthropic in general. We don't need to rehash it because we talked about on Sharp Tech last week. And that was one of those ones, I actually got a fair number of requests. Like you need to write this out, this, this idea of the sort of fusion of the safety culture with lots of choices that they make that are great for their business. And like, what a magical alchemy that is. And so I wrote that on, I wrote that on Monday. We already covered it last week. Go listen to that. But it was, you know, Fable was a way into that. And I said it there, I'm like, I'm not actually going to really talk about this because who knows what the heck is going on.
11:11 Ben Thompson? So you, but implicit in that, and anyone who's been listening to these podcasts for a while knows, is an overall frustration with Anthropic and a general concern about this company. Not because they're purposely being bad actors, but my overarching concern of people who think they are the only ones who are trustworthy and ought to make decisions for the rest of humanity is just bad.
11:43 Andrew Sharp? The track record is not great.
11:44 Ben Thompson? Yeah, I don't think that works. I don't think that works. And that's, and that's underlies so much about their sort of posture in general. I mean, there was like your write up of this AI summit or whatever that, that are a day or two ago. And you could just see in Amade's comments, it's like clearly he doesn't trust OpenAI. He's like, kind of trusts Google, but it's like, we should decide. And I think Altman took a much better sort of approach, which is actually, we shouldn't be deciding, someone else needs to be deciding. And I think that, you know, that's closer to my sensibilities, I would say, say for sure. So in general, frustrated with Anthropic, like not just in this case.
12:21 Andrew Sharp? Yeah.
12:22 Ben Thompson? At the same time, from what we've been able to garner, there's a good chance that actually the administration is just being incredibly stupid here. And so I almost feel like I've been a little too hard on Anthropic this week because from what I can gather, and again, who knows, you know, we don't have enough details of what's happening about what's not happening. But there's sort of an Occam's razor explanation here. Which is the US administration is like, stop it from doing this thing.
12:55 Andrew Sharp? Yeah.
12:56 Ben Thompson? And Anthropic is like, you can't stop it from doing this thing.
12:58 Andrew Sharp? That's not how AI works.
13:01 Ben Thompson? That's not how it works. And the, you know, the example here is what's the difference between find the vulnerability in this software versus help me patch vulnerabilities in this software? It's the exact same request. And the reality is Mythos is clearly, if you, again, if you used Fable, you could just feel it's clearly ahead. It's something new. No one does that. No one has anything to match it. But if the sort of history of the last few months slash years continues, we're going to have Mythos level models in the near future.
13:40 Andrew Sharp? Yeah.
13:41 Ben Thompson? And so what is the best possible thing we can do knowing that's coming? Everyone needs to be working overtime to patch their vulnerabilities. And we've been talking about this for several months on Sharp Tech. Like, it's going to be a messy few years. And it feels like there's this sense in the White House of trying to patch, you know, put your thumb in the dyke as far as this wave that's coming. And the reality is, is the wave is going to go over the whole wall. Like, it's going, it's going to be a mess. And the best thing we can do is offense is the best defense. Or I don't know if that's the right analogy here, but go out there, find stuff, work feverishly for the next little bit to fix as many vulnerabilities as we can
14:24 Ben Thompson? before this is truly broadly available. And every day that this conflict is going on are days when people aren't able to do this. And I think it comes back to, and this is where I do bring Anthropic back in the conversation, how many people in, there are some people in the White House for sure, that understand how AI works. But at the highest levels, this, this distinction between a probabilistic piece of software versus a deterministic one, deterministic when you put in hard rules, it's like.
14:53 Andrew Sharp? Yeah. Deterministic, you could say you can't make this available and this isn't possible. And you could just like hit some buttons to turn off different features that create unreasonable risks. You can't really do that with AI.
15:07 Ben Thompson? Right. That's like the magic of AI, right? So the strengths are weaknesses. That applies to sort of everything. The reason why AI can, is. You know, is it truly fully generalizable? I think not as much as sort of the advocates claim, but it's pretty darn generalizable. And the reason it is, is precisely because it's not deterministically designed to handle every sort of thing. And we see this in like self-driving cars, right? The, for, the reason why self-driving cars took forever to actually be viable is because people spent years trying to hard code every single scenario that a, a, a car might encounter.
15:43 Andrew Sharp? Yeah.
15:44 Ben Thompson? There are so many variables in a car that like you can't do that. What you need to do is you need to have an end-to-end model that is basically, it, it's not exactly the same as an LM, but it's the same concept of sort of a neural net that generalizes from being exposed to tons and tons and tons and tons of scenarios and develops its own sort of internal heuristics and probability distributions such that when something new happens, it kind of knows how to react without being told what to do. And you think about this, it's not just that it's not deterministic. It's that putting in deterministic rules into that functionality makes it worse in many respects. Because you're, you're, you're actually limiting it artificially in its capability to understand different situations.
16:29 Ben Thompson? And we're just moving into this fundamentally new paradigm that's totally different. And it's really easy for people to look at AI as a fancy search engine.
16:40 Andrew Sharp? Yeah.
16:40 Ben Thompson? We had this conversation a few years ago. You used AI for, for the greatest of all talk, looked up some stats.
16:46 Andrew Sharp? Which, by the way, it's gotten a lot better at pulling recent stats and everything else. It's a lot more accurate than it was two and a half years ago. But indeed, it's a different sort of tool than traditional Google.
17:00 Ben Thompson? Right. And what, what I told you at the time is like, no, that's the exact wrong function that you're trying to do, right?
17:06 Andrew Sharp? And so it's the worst thing you could use ChatGPT for.
17:09 Ben Thompson? Well, then this gets into also like, there's so many issues that, or so many concepts that flow out of this. Like what's a point that I've made again and again is actually a lot of this, you know, what is the shape of the crash going to look like or the bubble popping? It's gonna look like is people attempting to plaster on AI to existing workflows that were built around a certain way of computing. Being frustrated, it continually doesn't work. And then what's going to actually happen is people are going to completely rethink approaches. This idea of moving from putting advertising next to an article to a feed like.
17:48 Andrew Sharp? Yeah.
17:49 Ben Thompson? Like AI.
17:49 Andrew Sharp? Companies built atop this technology to take advantage of it.
17:53 Ben Thompson? That's right. And so this in the the challenge is is this takes years. It takes a long time for people to even reset their expectations of what's possible and what this means. And in the meantime, what we do have is this ability to look at code bases, find vulnerabilities and exploit them. Basically autonomously.
18:14 Andrew Sharp? Hack the planet, just like hackers, great movie. So to anchor people in this take, I think a lot of this springs from the report that. Amazon was able to jailbreak Fable and raise some concerns with the White House.
18:30 Ben Thompson? What this jailbreak is, right? There's this analyst researcher that saw a report from Anthropic, so this is all Anthropic spin, to be clear. Who characterized it as a very basic, like just slightly change your wording and now it's looking for vulnerabilities. Which I think, again, it's an Occam's razor explanation. It's actually very, you could believe that would happen. And then Anthropic saying, well, look, that's very minor. You can't stop that. And the White House being, you guys doing it again. We're just asking you to stop it and you won't stop it. And that's where Anthropic's at fault. They've created the conditions of mistrust with the administration.
19:10 Andrew Sharp? And also though, like, is the core issue for Anthropic? That the way they discuss AI risk then creates unreasonable expectations for how it can be managed?
19:21 Ben Thompson? The whole mythos rollout of this is like, like they go on for years talking about AI is like nuclear weapons. Then they have this huge project glass wing. We're going to like, this is way too dangerous to. We're gonna secure the banks and secure the big tech companies. And then like two months later, okay, everyone can use it now. And then you turn around and like you're creating the conditions.
19:46 Andrew Sharp? They create the expectation of perfect security, given how dangerous this weapon is for the whole world. And we're going to be very careful. And then when it's revealed that actually you can't possibly paper over every possible risk that arises from this technology. Other people who've been paying attention freak out and start to try to take back control. It seems like that's some of what's happened here.
20:10 Ben Thompson? Yeah, the problem is that this is very like, I'm not minimizing the risk. It's actually quite scary, like almost all software on the web is likely insecure. Just and if, but the problem is that you're not going to foreclose that forever.
20:27 Andrew Sharp? Entirely, no.
20:28 Ben Thompson? You're, you're, what is the administration buying with this? Weeks or months, maybe a year.
20:35 Andrew Sharp? Okay.
20:36 Ben Thompson? And the best way to solve the, the only way to solve the security problem is to go through it. Is to actually work and patch stuff like crazy. And by the way, the defenders have an advantage. They have the code. They can, they can examine their code bases. They can fix the problems. Even if you're attacking something from the outside, you're, it's a little bit more of a challenge. It's more obfuscated. You, you, you, you have to sort of figure it out. And the, the, the goal here needs to be to map like the Manhattan project we need is a whole scale. Let's go through all this code on the web, all these open source projects.
21:19 Ben Thompson? With to Anthropic's credit, they did do that, that's project glass wing sort of aspect. But you, we're, we're just not there. There's, there's, there's just not an acceptance of this reality.
21:32 Andrew Sharp? Yeah. Well, Michael says, if Ben's going to get so outraged about Anthropic's actions, even when he believes the safety threat is real, I'd love to hear a positive vision for what they should be doing to address safety concerns that wouldn't be self-serving. I agree with Ben's analysis about this being a convenient source of cultural alignment with the business incentives, but I just don't have a problem with it. What am I missing? So if you were running policy at Anthropic, what would you advise them to do to handle some of the safety risks?
22:05 Ben Thompson? The problem I have with Anthropic is the implicit and sometimes explicit, only we can handle this. Only we are responsible. Only we are properly concerned about the dangers. It's that exclusivity that is, and what's implicit in that is only we should control AI. Combined with the rhetoric of AI is going to run the world, people got mad at the sentence in my article, but it's just a logical construction. If A equals B and B equals C, then A equals C. I'm like, they want to run the world. Like, that's the implication of saying we should run AI and AI is going to run the world.
22:48 Andrew Sharp? Mm-hmm. I mean, how explicitly have they ever said we should run AI? Is it more inferring from their behavior?
22:55 Ben Thompson? I mean, yeah, what behavior? And I think their rhetoric, they clearly don't think Sam Altman should run AI, that's for sure.
23:01 Andrew Sharp? That's true.
23:02 Ben Thompson? Which, by the way, I completely agree, to be clear.
23:05 Andrew Sharp? And it sounds like, to Sam's credit, it sounds like even Sam agrees on that point at this, at this stage.
23:11 Ben Thompson? Well, he's cleaned up his rhetoric a lot. I think there's a, there was a real, a lot of bad rhetoric that came, and this is the challenge, because people are like, look, why should they not talk about like jobs upheaval? And no, there's the opposite mistake, which is to say nothing is going to happen, and where you're basically lying through your teeth. Like, and I think there's an aspect of that in Silicon Valley that is like, nothing, this is not going to, no, it is going to be disruptive. Now, we might not know exactly how it's going to
23:41 Ben Thompson? Be disruptive. There's no, you go back to the internet. The internet was incredibly disruptive. Like, the entire structure of society.
23:49 Andrew Sharp? Took 25 years. We were the frogs in the boiling water for a while there.
23:54 Ben Thompson? The administration is the Trump administration, which is directly downstream from the structural changes wrought by the internet. And what it did to media, like, so, the, the, the, you know, imagine going back to 1993 or whatever, when the world of web comes out and say, this is going to lead to Donald Trump being president, right? It's, uh, would have blown people's minds. AI is going to be arguably even bigger, but part of the problem is the, the certainty and prescriptiveness of what's going to happen. There's, there's an aspect of things are going to change and I think everyone could do with a little more, doubt, what's the word I'm looking for?
24:31 Andrew Sharp? Epistemic doubt?
24:32 Ben Thompson? Humility about like, you know, we don't know what's going to happen. This idea of just explaining why this is different. Appreciating that just because you understand AI and other people don't, doesn't necessarily, one, mean you're smarter than them, although often you are, to be fair. Uh, or two, that you know everything about everything because of this, this factor. Right? Like, it's hard to imagine, this is, you know, this is in my head, so this isn't like, anything, but it's hard to imagine, anthropic, you've been in conversations with people where they're explaining something to you.
25:15 Ben Thompson? And their tone and approach is so patronizing. And like, it's just like, it's just infuriating and you're like, screw this guy, I don't care.
25:27 Andrew Sharp? Are you describing conversations that you and I have had about various consumer technology that you demand that I adopt? And then I stubbornly refuse to?
25:36 Ben Thompson? Right, and you want to punch me in the face? Uh, so I guess, so maybe I can identify this attitude because I can personify it at times. Uh, that's a fair pushback. Uh, and I think the emailer does have his finger on something. It's, it's.
25:51 Ben Thompson? When I was at Microsoft, I remember I never got promoted.
25:57 Ben Thompson? And I was very frustrated about it. And like I was do, I ended up on this team where I was actually, I was level 61 and I was, everyone else on the team was level 63. And I was doing all the work and I'm like, well, you know, I should get, but my feedback was, the one piece of feedback I remember getting at my year review or something like that, was your what is excellent, your how needs a lot of work.
26:22 Ben Thompson? There was this, like, this idea of like, it's not just my manager that has to like be like the other managers on the promotion committee and the manager above the manager. And like all those, yo, there's an aspect of corporate politics that is just putting on a happy face and talking to people and selling yourself and your ideas. That this is why I didn't last long in corporate America. I'm much better off, I think, on my own, to say the least. And there is an aspect here to, my feedback to Anthropic is your what is fine, it's the how that needs work.
27:02 Ben Thompson? You need to figure out like how to communicate, how to communicate these things and the-
27:08 Andrew Sharp? And describe, like, there needs to be a capacity for risk tolerance over the next four or five years. And there's not much of that coming from Anthropic in the course of these conversations. They create the illusion of perfect security as long as they're the ones in charge of that security. And that's going to be a problem long-term. I'm open to the possibility that members of the Trump administration just simply don't understand that you can't keep these models from being jailbroken and you can't reduce the risks on that front. I will also note though, like, there was that report in the Washington Post about Anthropic giving the government a list of 111 organizations that had advanced access to Mythos.
27:52 Andrew Sharp? And then there was an additional 50 or so entities that were added and had already received access. And Anthropic was- Slow to get that list of companies to the government as the government was like, who actually has access to Mythos? And one of the entities was a South Korean company. And it just, what's interesting in that report is it seems like Anthropic still doesn't understand that the government is ultimately in charge and it's really not that complicated. Dealing with the government shouldn't be as impossible as Anthropic's making it look.
28:25 Ben Thompson? Yeah.
28:25 Andrew Sharp? So again, blame belongs on both sides.
28:28 Ben Thompson? This goes back to the Department of War debate, which ultimately was, is like a philosophical question of who's in charge. And the fundamental problem with Anthropic's how is actually not just the how, it's the substance. The, in that debate, their red line was, we get to decide how this is used. So as you asked earlier for evidence of them saying they should be in charge of AI, there's your example.
28:55 Andrew Sharp? Yeah.
28:56 Ben Thompson? Like they've already demonstrated it. And by the way, you know, I only talked about this last week. They've also demonstrated the capability and willingness to silently nerf their models to stop behavior that they don't think is acceptable.
29:10 Andrew Sharp? Yeah.
29:10 Ben Thompson? And so that, this is the core issue. The, there is an assertion of authority that is reserved solely for them. And it's not just with their customers. It's not just with other companies. It's with the federal government. That is the fundamental issue. So we're, it actually does go back to that episode. And that's the issue here is they're trying to dictate terms and the government is probably being dumb. The like export control thing, who knows if it's even legal or whatever it might be. That will, that's for like the courts to decide. I, my guess is probably not, right? But.
29:54 Ben Thompson? What's actually happening is the exact same battle that was happening before. Who gets final say?
30:01 Andrew Sharp? And when the government says jump, Anthropic should say how high? You want the additional 50 entities? Here are the additional 50 entities. You have a problem with the South Korean entity?
30:12 Ben Thompson? How's that boot taste, Andrew?
30:12 Andrew Sharp? Well, no, that's just the way it's gonna work.
30:17 Ben Thompson? No, that's the thing.
30:17 Andrew Sharp? It's inevitable. If you're talking about a super weapon that poses dangers to the entire world, like, this is the inevitable conclusion of that conversation. But they still have not fully gropped that.
30:28 Ben Thompson? And that's what's so frustrating is the people who are implicitly saying they should decide everything seem incapable of understanding the reality of the world as it actually is. And if you can't understand the reality of the world as it actually is, on what basis should I have faith that you get to decide how the world's going to be?
30:50 Andrew Sharp? All right, and that is the end of the free preview. If you'd like to hear more from Ben and I, there are links to subscribe in the show notes, or you can also go to sharp tech.fm. Either option will get you access to a personalized feed that has all the shows we do every week, plus lots more great content from Stratechery and the Stratechery Plus bundle. Check it out, and if you've got feedback, please email us at email at sharp tech.fm.
Previous (Preview) Five Questions on WWDC 2026, Fable 5 And Its Guardrails, What Anthropic Has in Common With Apple Next (Preview) A Summer Break Mailbag: Memory Mania, Vibe Coding, Mafia PR, Caffeine Intake, Garages, and How to Fix Soccer Page updated October 30, 2026 · transcript and summary are machine-generated
