← Sharp Tech with Ben Thompson
Sharp Tech with Ben Thompson
(Preview) The Anthropic Mess Continues, Frontier AI and the Uncertain Future of Law, Q&A on Netflix, Dating Apps, F1
00:29:25
Loading summary
Transcript89 lines
- [00:00]
A
Hello and welcome to a free preview of Sharp Tech. Hello and welcome back to another episode of Sharp Tech. I'm Andrew Sharp and on the other line, Ben Thompson. Ben, it's great to see you. It's great to be back. How you doing?
- [00:21]
B
Doing okay. I mean it looks like you just leave, go on vacation, let me go to my own devices and I just go out and get canceled apparently.
- [00:29]
A
Can I tell you what, Nothing on my vacation was more relaxing than not recording a Friday night emergency episode about Anthropic versus Oh yeah, we're on emergency episodes.
- [00:41]
B
Yeah, yeah, we're not doing those anymore.
- [00:43]
A
We may be post emergency episode on this podcast, but we are going to talk about Anthropic and the Trump administration on this podcast. One week ago, for anybody who's missed it, the Trump administration canceled a defense contract with Anthropic and announced a plan to phase Anthropic out of all government government institutions within the next six months or so. That will take some time. Secretary Pete Hegseth if it happens. Yeah, TBD on all of this. It's changing by the day here. Pete Hegseth also threatened to label the company a supply chain risk to national security, potentially implicating many more Anthropic contracts.
- [01:21]
B
Well, he went beyond that. He said that no one who has contracts with the US can work with Anthropic. Which goes even supply chain risk is
- [01:31]
A
a very serious designation. Yeah.
- [01:33]
B
It also does not go that far. So there's like there, there's, there's been some definitional issues all along broadening the
- [01:41]
A
scope of the potential implications, at least in Hegsis tweet and they have not followed through on that threat one week later. So again, TBD on a lot of this. For now. We will talk about everything you wrote earlier in the week. I wrote about it for Friday as well. We do have to start with the updates as we record Thursday afternoon here I'll read from the information which republished an internal memo from Dario Amadei to Anthropic employees. It was a long memo. This is not the whole memo, but it did culminate with this. The real reasons Dow and the Trump administration do not like us is we haven't donated to Trump while OpenAI slash Greg have donated a lot. That's Greg Brockman for anyone keeping track. Dario continues and says we haven't given dictator style praise to Trump while Sam has. We have supported AI regulation which is against their agenda. We've told the truth about a number of AI policy issues like job displacement and we've actually held our red lines with integrity rather than colluding with them to produce, quote, safety theater for the benefit of employees. In in parentheses he says, which I absolutely swear to you is literally what everyone at Dow Palantir, our political consultants, etc. Assumed was the problem we were trying to solve. Sam is now, with the help of D O W, trying to spin this as we were unreasonable, we didn't engage in a good way, we were less flexible, etc. I want people to recognize this as the gaslighting it is and then just to keep all of us on our toes, like four hours after that memo leaked, we got a headline from the Financial Times anthropic Chief back in talks with the Pentagon about an AI deal. So what do you make of that?
- [03:32]
B
Important. What important detail about this memo is this was a memo he wrote last Friday. It was not heat of the moment. He wrote yesterday. Yes. So the memo was published on Wednesday, but yes, it was in the heat of the moment. Is the memo a good look? No, it's not. Does the memo evince a accurate theory of mind for everyone that Amadei is interacting with? I don't think so. I found this memo quite validating actually in terms of what I wrote, which it's always the you know, I have a principle. Assume everyone is smart and figure out why they're doing dumb things. And no one is denying that Darrell Amadeh is smart and brilliant and everyone at Anthropic is. And yet I've felt for ages they and this sort of broader movement has had a significant blind spot, which we talked about in our first podcast of the year when we were going to the dorm room. And that blind spot is this entire article we talked about that Dorkish Patel wrote about the future when AI does everything and humans have no worth. Everything in the world was different except for modern conceptions of property rights, which were still the exact same. Yeah, and like that was one of the pushbacks we had and you almost feel silly pointing this out. And we're going to get to. Sorry, a couple emailers. I know I'm ready to go in on you for similar reasons here. But like there's just this weird blind spot that is consistently there that I feel like was at role in this case. It's like there's assumption that certain rules and laws are immutable and no sort of conception about how do laws come to exist. Like it's kind of funny because the first article or when ChatGPT came out, I wrote that an article called AI homework. Yeah. And just sort of like, you know, it sort of opened with a. I asked a question about Hobbs versus Luck, and the chatgpt got it wrong. And I thought that was sort of interesting. Sort of started the initial, like, hallucinations, obviously a bad thing, but also amazing. And then you can sort of understand how that particular hallucination happened because Hobbes and Locke are always referenced together in, like, textbooks, and it basically confused their two views. And this is shafty B.3.5.
- [06:07]
A
Right.
- [06:08]
B
This was very fine in that debate.
- [06:10]
A
One of my favorite articles of yours.
- [06:12]
B
Well, thank you. I think that's a compliment. But maybe there was a bit. Where can we all go back to Hobbes and Locke, or just political philosophy in general? Where do rights come from? Like, what is the foundation of authority and government? And Hobbes was sort of like the monarch, and Locke was more on, like, the democratic sort of will the people. But even then, it goes back to what's the foundation of rights? You know, Totally.
- [06:44]
A
I mean, I noted that in my article. It's like the idea of property rights, which are supposedly implicated by some of what the US government is trying to do to anthropic Those themselves are a legal construct that is given force by the American government that allegedly imperils the property rights.
- [07:03]
B
Exactly. And so when I read. When I read this memoir, it's like. It's interesting because I think one of the frustrations I had when I wrote what I wrote, which I didn't expect the degree of blowback that it got, but that's fine. I stand by it. It. Actually, I'm gonna bring up the blowback a little bit because there's an analogy to a previous article I wrote when we talk about your piece in a little bit. But you certainly have people who only see things through a partisan lens, which is very frustrating. And I. The whole point here is this is not a partisan issue. I think it wouldn't matter who's in the White House. Maybe today it matters, but, like, in 10 years, if AI is as powerful as it potentially is, it's not going to matter. It's going to be the government versus the AI companies. And there's a bit where I'm actually fairly grateful. This came up in the interview I had with Gregory Allen this week, which I highly suggest everyone listen to. I thought it was really, really good. And it sort of like fleshing out some of these issues which he doesn't completely agree with me with. Like, we went back and forth on a few different things. But there's a bit where it's. I'm quite grateful this debate is happening right now in 2026, when, all things considered, the stakes are pretty low because there's a potential that the stakes are drastically higher in a very short amount of time given the exponential increase in the power of AI. And there's actual real fundamental questions and at play here that need to sort of be discussed. And so I don't think it's a partisan issue. And yet, of course, there's the category of people for whom it's a partisan issue.
- [08:39]
A
Right.
- [08:39]
B
And what's frustrating is they don't even attempt to grapple with the article. It's just say, like, is this broadly in line with the Trump administration or not? And then they get insulting and call me a fascist and a bootlicker and sort of, whatever. It might be sort of easy to dismiss.
- [08:54]
A
You're only. What was a liquor when we're talking about antitrust. Just to clarify for the audience out there.
- [08:59]
B
That's right. I'm a bootlooker for the corporate. Our corporate overlords, not our political ones. So. So I was going to clip that. So it was a risk on my part.
- [09:09]
A
But here we are, you know. No, I mean.
- [09:12]
B
But I was frustrated about the response was from the. Again, we're going to. For just categorization purposes, we already have an issue with the Department of War versus Department of Defense. Anthropic use Department of War. So we'll use Department of War.
- [09:27]
A
Not a party for everybody out there. I believe there is some sort of congressional. There's a congressional threshold that needs to be cleared before it's officially the Department of Defense. But everyone involved in this mess is referring to it as the Department of War. It is a bit of a minefield in D.C. depending on who you're talking to. You have to choose your nomenclature carefully. But Department of War for the next hour here, as we talk through all
- [09:53]
B
this, I was just using the part of work because anthropic did. I wasn't really thinking about it, but I just sort of assumed that was the case. And then some people were coming at me, oh, you're saying department. I'm like, honestly, I did not give. So I've just been saying the US Government basically ever since. But whatever it might be, I might try, like that first sentence in my article or the quote, you might not care about politics, but politics cares about you, or whatever it is, that wasn't just anthropic, that was also about me. So to be Clear. What I found disappointing was the other sort of categorization that I'm going to use in this podcast for ease of use is sort of ea, the effective altruists, out of which a lot of the AI researchers come. Anthropic is kind of the standard bearer as far as AI companies for that. Now there's a gazillion definitions in who it is or whatever it might be. This is a very blunt instrument. I'm just using it for terminology purposes.
- [10:50]
A
Okay.
- [10:51]
B
But I think one of the frustrations I had with the EA critiques is they sounded a lot like the Trump critique, like the anti Trump critiques, where it's like, did you actually read the article? Instead there was just sort of visceral reaction that seemed to like I addressed that. That was in there, like whatever it might be.
- [11:16]
A
So how so what were they objecting to in terms of the EA side of things?
- [11:21]
B
Oh, well, no, actually a lot of them just fell into Trump. Like things like, oh, he's trying to back into a conclusion here because he wants to support sort of xyz. And a lot of them really fell into this property rights will last forever sort of categorization. When my whole point here is laws ultimately are human constructs and they come back at the end of the day. Laws are formed. Nations are founded on the tip of a bayonet. Go back to Hobbs, Hobbs and Locke.
- [11:56]
A
Here we are.
- [11:57]
B
That's right. Hobbes in particular. Right. And given that a lot of people. This is where I brought up the Palmer Lucky. We should let democratic leaders decide because ultimately they have the legitimacy. I was not making that argument.
- [12:14]
A
Okay.
- [12:14]
B
Because what I was talking about was the risk that we face here is that AI becomes so powerful, it becomes a source of power in its own right. It becomes the ultimate gun, the sharpest of the bayonets. And when that's the case, the problem is the call it temptation or call it felt necessity of the government is going to be to violate property rights to leverage their monopoly on violence. Not just because they want the technology, but they actually fear they're losing their monopoly. Yeah. And that's anti democratic. That's why I'm not relying. My point was not a democratic argument. My point is the risk we are running if we don't get alignment here between AI companies and the government is we're creating the conditions for the government to turn anti democratic, to violate the laws. The laws exist and are enforceable because the government enforces them. What happens when the enforcer is the one that decides the laws don't matter and need to be violated.
- [13:26]
A
Right. We're seeing.
- [13:27]
B
Who do you call to. Right. Are you going to call the international police? Right. And so this is actually, I kind of wish I would have made this more clearly. I think I did in that Gregory Allen interview, sort of like fleshing this out. And I almost regret even bringing in the Palmer Luckey bit because I did say in there, this is the big reading. I'm like, this is not my argument.
- [13:47]
A
Yeah.
- [13:47]
B
What I should have added is it's not just not my argument, it's that my concern. And it's so funny. People were calling me fascist. My concern is fascism. We're creating the conditions for the government to violate its own laws, to obliterate the concept of property rights precisely because they fear the power of AI more than they fear the ultimately human constructs of laws and the self control to abide by them.
- [14:19]
A
So let me tease out what you're saying there. Are you saying that you are free? Fear is that if there isn't close coordination with all of the frontier AI labs, that these labs will then be nationalized and weaponized by the state without any sort of input from those labs.
- [14:37]
B
Yeah, that's. And that's arguably an optimistic outcome.
- [14:40]
A
Okay, like so like sort of a techno dystopian hellscape that could emerge if the government there's.
- [14:47]
B
Yeah, yeah. So like just at a very fundamental level, the government has a monopoly on violence. You know, there's a group. Schuber brought this up in dithering this sort. I think it was. He said it was P.J. o'. Rourke. I wasn't familiar with the article, but the idea that at the end of the day you pay taxes because the government has guns.
- [15:03]
A
Yeah.
- [15:04]
B
And it's like none of us think that way. We just pay taxes because that's what you're supposed to do. But ultimately what happens if you don't pay taxes?
- [15:12]
A
They come and take your house and you go to jail.
- [15:15]
B
And who comes to take you to jail?
- [15:17]
A
Probably some people with a gun.
- [15:19]
B
Right. This is why I thought the Iran. That's why I started with the bit of international law in Iran. Sort of a useful analogy, I thought
- [15:29]
A
like a sensitive issue on Monday morning as everyone's processing the world. I know probably did, you know, as far as the backlash is concerned.
- [15:38]
B
But yeah, but so the issue, the issue isn't just that so the government has a monopoly on violence. The problem is if AI is what it is promised to be, it's that the government doesn't just want it for its own purposes. It's that it will feel threatened by it.
- [15:57]
A
Yeah.
- [15:58]
B
And again, is it threatened by it today in 2026? No. If these exponential curves actually play out, is this potentially going to happen sooner than we think? Yes. And so if we want to retain our system of government, if we want to have democratic control, we ought to, all of us as an industry, be thinking about how do we not create the conditions that the government feels oppositional to and threatened by AI because that last week the government felt oppositional and threatened by AI and fortunately, we're still in a place where we can laugh at it and say they overreacted, they overdid it.
- [16:44]
A
Yeah.
- [16:44]
B
But what if AI was 100 times or a thousand times more capable and powerful than it is today? And that might be here pretty soon.
- [16:53]
A
Indeed. Well, before we get to more about what you wrote last week or this week, excuse me, two questions on Anthropic's next moves here. First, Scott sent this in after the memo leaked. Subject line was strategy credit overdraft. He writes hello, gents. Or no, he wrote hello, gents. One of Ben's hottest takes recently was that his concept of a strategy credit is his greatest idea, even better than aggregation theory. I can't disagree. Is it possible, though, to exhaust one's credit or overplay the strategy credit hand? I wonder if Anthropic has a strategy credit in saying no to the Department of War to further sell themselves as the ethical AI shop. Better to attract talent and consumers. But could this strategy credit lead them so far in one direction that that they undermine their core business? It feels a bit like a politician winning the primary, but going so far left or right that they can't win the general election. What do you think about that possibility?
- [17:59]
B
Yes, but I would actually. I don't think it undermines their core business. I think it probably by and large helps their core business. Number one, you have the Pentagon saying, we're going to ban you because we watch you so much and you won't help us. Your technology is so awesome. You have the talent wars where I'm sure by and large across AI talent, people are much more sympathetic to and align with Anthropic's views.
- [18:24]
A
Yeah, I would guess they'll gain more talent than lose anything in the next couple years.
- [18:29]
B
You have this, you know, they've been number one in the app store, I think all week. Right. Like in people deleting OpenAI going with quad and maybe for the first time and maybe it turned out those super bowl ads, which I thought were somewhat deceptive, but whatever, all ties together and they're building this sort of this great brand and the. So the problem here isn't that they're undermining their core business. The problem I have is they are becoming unaligned with our democratic government.
- [18:58]
A
You know, I mean that does undermine their core business. That's the risk that I see as far as the next year or two for Anthropic here.
- [19:07]
B
My worries are much more basic than that. I'm worried about them undermining our democratic government and way of living. And the means of that isn't necessarily anthropic acting badly, it's prompting, putting in the incentives for the government to act badly. And again, like the great irony about this is, particularly on the domestic surveillance point, I have tremendous sympathy and by and large am aligned with Anthropic. I think the holes we have, the loopholes we have for domestic surveillance are fairly atrocious. I think a lot of our privacy policies were predicated on a world where there was friction, it was hard to spy on everybody. And we've already like blown past them to a great degree. Cause computers can wiretap everyone. Like just like one of those secret closets in the AT&T building. Don't ask. You know, and now we can analyze all that sort of at tremendous scale. I think it's a totally legitimate concern. But everything is a trade off. And I just don't think this was a fight worth picking particularly. And this is where we don't know what the details were. You can't have, you just, it's not tenable to have a situation where on a case by case basis you sort of get to have a say. Let's get as clear of guidelines as we can and that the lawyers in the DoD DoW can decide, go, no go. And they don't have to ask you for permission. And let's do that not just because we support national security, but let's do that because we are developing potentially the most powerful technology ever.
- [20:48]
A
Yeah.
- [20:49]
B
And we need to appreciate the threat that is going to pose to not just people's jobs, but not just SaaS companies, but to the very fundamental legitimacy of the United States of America.
- [21:05]
A
Indeed.
- [21:05]
B
And, and like this, actually the other funny criticism I got was that, oh yeah, Ben just wrote that because he doesn't like take seriously AGI or like the power. Like that's the exact opposite. It's because I'm taking you seriously.
- [21:22]
A
Right.
- [21:22]
B
Like if I was convinced, I can see the arguments that this is just another technology, just like computers. It's actually not gonna be that big of a deal. In which case it's like, yeah, let's unapologetically support them because just we saw abuse that happened with computers. But I am taking seriously the possibility this is unbelievably powerful and destabilizing. And for that exact reason, it's super important that we be aligned with our democratic system of government.
- [21:50]
A
Well, related to this. One more question on Anthropic's calculus in the short term here. Adam says strip down analysis in your post on Monday is pretty banal. Anthropic can't and shouldn't be able to dictate its ToS with the US government because of obvious power dynamics. Here's the thing. No firm in Anthropic's position would be unaware of the US Government's ability to force Anthropic's hand. Forget the supply chain risk designation for a minute. The real hammer the US Lords over Anthropic is the Defense Production Act. It's not quite all the marbles, but it's most of them. So why then has Anthropic performed such very public line drawing over the last week? What's the end game? Any guesses on the end game and or why this is all played out in public the way it has? Because I don't think that has been strategically optimal for Anthropic.
- [22:43]
B
I don't know. I can speculate, but I'm hesitant to. I think one of the critiques that can be aligned at maybe the EA movement to a bit is and where I do feel a certain sort of allergic reaction to them is this sort of we are the smartest people in the world and we're the only ones that truly think through costs and benefits.
- [23:10]
A
Yeah.
- [23:10]
B
Everyone else is kind of stupid and that's one thing, but it sort of follows on. Therefore, we should make decisions for everyone. And one of the big issues I've had with the AI labs in general, and this goes back to OpenAI's being a nonprofit, certainly applies to Anthropic is this sort of assumption that AI must be used for the good of everyone.
- [23:35]
A
Yeah.
- [23:35]
B
And the question is, who decides? And when you push on it, it's always we do. And I am very allergic to that point of view. I think that leads to hubris. It leads to very dystopian outcomes. It leads to the worst abuses we've seen in human history. People don't. The worst actors are not the ones that are attempting to be Evil. It's the ones who are so convinced that they are good that everyone who opposes them ought to be eliminated for the greater good. Yeah, I'm not saying, I'm not saying they're there. I'm just saying, like, that's, like it's shades of that.
- [24:15]
A
Well, I can see your point in the memo. There's a fervor permeating that memo. That, that's why I said the memo
- [24:24]
B
maybe felt justified because, like, this is, this is not fully grappling with reality. Like, you know, is Sam a tricky guy in an opera? Yeah, I think we figured that out. But there is a. There's a. Yes, a fervor. A fervor here. And I find fervors fairly terrifying. And particularly when the fervor is being evinced by the leader of again, potentially the most powerful technology ever. That's just unsettling. I find it very unsettling. And I do want to note, I thought the analysis is pretty banal also. Right. Like, and the, you know, this didn't rise nearly to the level of. But I would say this is like there was at least a shadow of the out of pocket response. I got to net neutrality in like the first year of the Trump administration when I've sort of told this story. I wrote about net neutrality in the context of Netflix versus Comcast in like 2014 or something. And I basically said Netflix is full of BS here in terms of like their peering agreements and stuff like that. I basically made the same argument in 2017 when the Trump. So Obama, and I think it was 2016, last year imposed Title 2 of the Telecommunications act to Internet data, which imposed net neutrality, but with this massive amount of bureaucratic paperwork that basically was designed to apply to AT&T back when they did long distance.
- [26:01]
A
Right.
- [26:02]
B
And my point in that article is that I'm definitely pro net neutrality, but this is the worst possible way to do it and it's going to restrict investment in broadband and blah, blah, blah, blah. A very technocratic article that I thought was quite banal and for which I not only got. Again, I got a few hundred replies here, I got thousands, tens of thousands of replies. I got death threats. I got like the whole sort of kit and caboodle. I had people like, who this, like take Neelay Patel, who this week called me a fascist on blue sky at that time, writing entire screens about how I don't care about democracy, like this whole sort of thing. And it's funny if you go back to that article, I quoted several Tweets, they're all deleted. Because every single person who lost their minds about that was totally wrong.
- [26:50]
A
Yeah.
- [26:51]
B
And not just that the vast majority of them changed their position in terms of wanting more control of they didn't want neutrality, Suddenly they wanted platforms making decisions about what was and wasn't allowed.
- [27:04]
A
Yeah.
- [27:04]
B
And I appreciated your piece this week on Sharp Text, which didn't get into the issues as much as saying these reactions are out of control.
- [27:16]
A
It's pretty exhausting. You know what the common thread was between your piece on net neutrality, which I have said in the past I cited in law school, got an A plus in the class on the final exam, cited Ben Thompson before I even knew you at that point.
- [27:32]
B
That's right. We never even met.
- [27:34]
A
But the common thread, I believe. Didn't Ajit Pai share that piece on Twitter?
- [27:40]
B
Yes, Yes.
- [27:42]
A
A member of Trump administration.
- [27:43]
B
So I deserve blame because I, well, I originally entitled it Ygpai is right Great. That title, that was a mistake. The title is no longer that.
- [27:52]
A
It is a fun, spicy title.
- [27:54]
B
Grow Net neutrality anti article 2 or something like very dry like that. You wonder why my title tend be to be fairly drunk. Yeah, yeah. But yes. And then Mo Michael retweets my article on Monday. Not helpful. I, I like I'm not on aide here. I'm just trying to make a point but that's right.
- [28:12]
A
Yeah. Well. And as far as anthropics end game, all I will say is that I wrote in my piece I would expect them to broker some sort of compromise in the midst of this.
- [28:23]
B
I think that's gonna happen. I, I, it's possible this happens even before this podcast publishes. Like I think the he was very, I think he settled down. He was at the Morgan Stanley TMT conference and I think was fairly conciliatory in his tone. Yeah, I'm hopeful that maybe in some small way what I wrote is like a little bit of like, like let's wake up here. Like this is, let's not go down this route.
- [28:50]
A
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 SharpTech. 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 STRI and the STRI plus bundle. Check it out and if you've got feedback, please email us at. Email sharptech FM.