Chapters
Shownotes Transcript
This podcast is supported by KPMG. Your task as a visionary leader is simple. Harness the power of AI. Shape the future of business. Oh, and do it before anyone else does without leaving people behind or running into unforeseen risks. Simple, right? KPMG's got you. Helping you lead a people-powered transformation that accelerates AI's value with confidence. How's that for a vision? Learn more at www.kpmg.us.ai.
I'll tell you something that went on for me this week. For a long time, I've used the Google Docs product. Have you used this product? The Google Docs? Yeah. You've seen this one? You've seen this one? Okay. So— A little obscure, buried, you know, feature. We like to highlight up-and-coming apps. So I'm in the Google Docs product yesterday, and I'm working with some folks for a project I'm excited about. And these people are Google Docs masters, okay? I didn't even know that Google Docs could be mastered because it looks so basic.
But they keep saying things like, we'll just go down to where I am or go up to where I am. And I was like, bro, I don't know where you are on the dock. Okay. Like I don't have some sort of like dock ESP. You can click the circle. You can click the circle. Did you knew you could click the circle? Yes. Okay. I have to tell people this. If you are in a Google doc with other people and you're editing collaboratively,
you can just click the little circle at the top of the page with their little initial or their picture, and it will take you to wherever they are in the document. Wow. I can't believe you're just learning about this. How long would you say you've known this? At least five years. There's no way. Yes. Five years. Why has this never come up before? This is insane to me.
You know, you know why it is? Cause it's in the New York times. There's like five bylines at every story and six editors. And so you actually have to know, like I work for normal media organizations where like, you're lucky if you could like get an editor to like come into the newsroom to help you out. And so I, maybe I, there just haven't been that many people in my Google docs. Wow. Casey, I'm going to blow your mind. Okay. Pro tip. You can print a document on Google docs. Now we're just saying things that no millennial would ever do, but I think it's great that Google has added these features. Um,
Generally speaking, you don't want other people in your Google Doc, and I'll tell you why, because they're going to have ideas, and those ideas are going to make more work for you. But if you find yourself in this sort of nightmare situation of working with other people, you can click the circle. Wow. I love the idea of a podcast where every week we just teach you how to use one extremely basic Office product. You're laughing, but I think that that could actually be quite useful to our readers. Next week on Artfora, Casey learns Excel. We're going to need more than one episode for that.
I'm Kevin Ruse, tech columnist at The New York Times. I'm Casey Newton from Platformer. And you're listening to Hard Fork. This week, how a private Discord server ended up at the center of a national security crisis. Then, more chaos and clownery at Twitter. And finally, what happened when researchers gave AI memories? I'm guessing it was nothing good. We'll find out. All right, Kevin, can we talk about national security for a minute? Absolutely. All right. Is this a secure line? No.
Yes, I am in the sensitive room inside my house where I handle classified information, so we are good to talk about this stuff. Let me tell you, every room in my house is a sensitive room depending on the day. But what I want to talk about is the national security aspects of it all. Yeah, so this is a big military intelligence story that is currently making front page headlines in the New York Times and other publications.
So on Thursday, federal investigators arrested a man named Jack Teixeira, who is a 21-year-old Air National Guardsman. And they believe that he is linked to this leak of classified military intelligence documents. These documents have been circulating on the internet for a while now, and it wasn't clear who had leaked them or why. But
these documents really did cause a stir. They've compromised some of our relationships with other countries. They've exposed weaknesses in the Ukrainian military. This is a big deal, and it seems to have stemmed from a very strange, very online source. So before we talk about the online piece of this, let's just talk about what these documents were and why it's important that they're out there.
there. Yeah. So these documents were, one, just very politically sensitive. Two, this was also supposed to be top secret. And there is stuff in there that the United States was trying to hide both from its adversaries and allies alike. It included things like details about secret American and NATO plans for building up the Ukrainian military, for example. And then,
Third, these are just timely documents, right? Sometimes when we hear about these leaks, they're years old. Somebody finds them in a file cabinet and blows a whistle. The information in these leaked documents is in some cases weeks old. And it seems like stuff has been leaking for even longer than that. So that's just a really big deal. Yeah. And I would say like,
This leak of classified information has really shaken the national security apparatus. So, you know, military intelligence officials have been looking into it furiously, trying to figure out where it came from and who did it. The State Department and other agencies are trying to handle some of the geopolitical fallout.
And Merrick Garland, the United States Attorney General, made a statement on Thursday about the case. That's right. And what's really interesting to us is how these documents have been circulated. These documents were found on Discord servers, and not just any Discord servers, but video game Discord servers. And at least at first, it seemed like people were posting these documents not to blow the
arguments online. And so, Kevin, when you and I heard that, we said, well, we need to get to the bottom of this. Totally. And before we get into the details of these documents and why they're being shared inside these peculiar Discord servers, let's just explain for people who maybe don't know what Discord is. Yeah. So Discord is a communications app. It
It's free. And while it started as something that gamers primarily used, now basically everyone's use it. And there are discords devoted to posting memes. There's discords devoted to making generative AI. I have a discord for my newsletter where people who read my newsletter can just kind of hop in and talk about today's issue and other things that are on their mind. So it's just kind of a very fast growing, very popular next generation social network, I would say, that is used by all sorts of folks. Right.
Right. So I would compare it to something like Slack or Microsoft Teams, but it's not work-focused. It started, as you said, as a gaming thing. And there are different kinds of Discord servers, as they're called, right? So anyone can start a Discord server. And some Discord servers just have a handful of people in them. But I've been in ones where there are also millions of people. So it can range from very small group chats to very large message board type things. Yeah. And
An interesting thing about Discord is that you typically belong to both public and private Discords at once. So you might have a very small, essentially a group chat with your friends, and then you also belong to a bunch of larger servers. And this winds up playing a key role in the plot here because what it seems like happened was that some materials that were designed to be shared only with a group chat pretty quickly migrated to some much bigger servers.
Yeah, so we are still piecing together exactly what happened with this leak of classified information. But what we do know is that the leak appears to have started on a private Discord server that was populated by a group of friends who played video games together. The server at one point was called Thug Shaker Central, which is, do you get that reference? I don't. Yes, I looked it up and I would say it is a meme that is not great. Yeah.
Like how not great. Like, I would say that this meme originated in the world of pornography. And I would say pornography with a somewhat racist bent. Oh, okay. So we won't talk about the origination of the name Thug Shaker Central, but that was the name of this Discord server. A bunch of friends are in there talking about video games, presumably talking about lots of other things. And...
At some point, someone in this Discord server, who we now believe is the person who was arrested, the airman Jack Teixeira, shares these classified documents in this essentially group chat. And this might have just remained in this group chat. But because this is the internet, because this is Discord, because, as you mentioned, lots of people belong to multiple Discord servers, things just got really weird and out of control.
Yeah. And look, I'm sure you're in many group chats, Kevin. I know our listeners are too. And if there is one lesson you can take away from this story, it is that stuff is actually going to get outside of the group chat. Yeah. What happens in the group chat doesn't stay in the group chat. And we don't know exactly what happened in this case, but it appears that some of this classified material was eventually taken from this Discord server, Thug Shaker Central, and shared with
to a public Discord server that was devoted to a niche YouTuber called WowMow. Have you heard of WowMow? I have now heard of WowMow. I've heard of WowMow now. Yeah.
But before now, had not heard of WowMow. How now, WowMow. And this is the thing, like lots of YouTubers have their own associated Discord servers where their fans will go and talk about their new videos or just talk about whatever. It's sort of this fan community. So...
these classified military intelligence documents end up in this niche YouTube Discord server for WowMao. And then from there, it seems like somebody in that server saw those documents, saved them, and then re-uploaded them to another Discord server. This one...
a big one devoted to Minecraft. And this really is the funniest part of the story because what appears happened is that in the midst of a minor spat, one of the people who had access to these documents just threw them into the chat by saying, here, have some leaked documents, which I think we can now agree is just a funny thing that we will now start saying to each other all the time.
It is funny, but it's also like, this is a serious leak, right? This is leading every major news organization right now. The State Department and other agencies are scrambling to contain the geopolitical fallout from this. So this is a major event. And the fact that it has such a weird and sort of hyper online origin story just made it totally fascinating to talk about. And so...
I thought we should bring in one of the reporters who has been really leading the charge on this story. And that person is Eric Toller. Eric is a reporter for Bellingcat, which is a group that specializes in what they call open source research, or just basically using publicly available internet data to kind of solve mysteries and do reporting.
He teamed up with reporters on the visual investigations team at The New York Times. And he is an incredibly good internet sleuth. And he was actually one of the reporters who broke this story in The New York Times. So I thought we should get him in and talk to him about where this leak came from and what he can tell us about how this strange online discord culture has birthed a true national security scandal. ♪♪
Eric Toller, welcome to Hard Fork. Hi there. So you are part of a team of reporters that have been digging into this Discord leak and just published a story in which the name of the alleged leaker or the leader of the Discord server where these groups were originally leaked is this 21-year-old from the Massachusetts Air National Guard named Jack Teixeira.
So my first question is, how did you and your colleagues track this person down? Yeah, it's a little convoluted. And there's a lot of different platforms that most people probably know about, maybe not everyone. So this whole thing came to a head and everyone realized what was going on. And the Times published a story about how the Pentagon was looking into this leak. And at the time, it mentioned that there were some posts on the messaging app Telegram. And so I was looking back like, you know, there's no way this is the original source. They had to come from somewhere. And I looked and looked and looked and I found...
three photos and two other ones that weren't in the Telegram cache on 4chan. But four of the ones on Telegram weren't on
4chan. So clearly they got it from the same place. I just don't know where that place is. So I looked and looked and looked, couldn't find anything. And then somebody DM'd me of like, hey, I think I saw this on Discord. And they pointed me to a Minecraft map server. It's where people make maps of Earth. They try to make real-life Earth on Minecraft. And this guy had posted 10 on there. And of the 7 that were posted on 4chan slash Telegram, those 7 were from this cache of 10. And there were 3 additional ones in here that no one had seen before, as far as I know.
So I talked to that Minecraft guy, very nice Minecraft guy. And he's like, no, it's not me because he was terrified for good reason. And he's like, no, no, he's also a kid. By the way, this is going to be a recurring theme. They're all kids, a bunch of 17-year-olds. And this kid was like, no, I'm not the leaker. Everyone, no, please, everyone leave me alone. I'm innocent. I got this from a different server, a different guy. And he got them from another 17-year-old, another kid who posted them on another server, but that's this YouTuber guy.
Yeah, so you're, of course, talking about the WowMouse server. Yes. So from there, I talked to a few other people, four people in total from the server. And eventually I figured out kind of from these people I was chatting to that it was a different server that this kid was a part of.
that he pulled from and then posted them on the WowMile. So there were hundreds and hundreds on this other server called... It had a lot of different names, but the name at the time of the leak was Thug Shaker Central, which is this meme from a couple years ago. And this is where hundreds and hundreds of these were posted on here. The 17-year-old kid took 107 of these and cross-posted them onto the WowMile server. And they've been being posted on this older server, this Thug Shaker server, since the latest...
October, it could have been earlier, but the latest around October of last year. So you followed the sort of trail of breadcrumbs back to this server, Thug Shaker Central. Yeah. And then from there... I think you'd say that out loud before. And then how did you figure out who in that server had probably leaked these documents? Yeah. And from there, I was like, okay, well, there were only about 20 or so active members on that server.
So the pool is pretty small. And I know, because I talked to the kid who also was 17, and he told me about, like, yeah, so he called him OG. So he, that's not what anyone else called him, by the way. Just because I was like the original leaker. He's like, well, let's just call him OG. Like the original guy or the original gangster or whatever. Yeah.
So this 17-year-old, this literal 17-year-old who's in this Discord server called Thug Shaker Central tells you that the person who originally posted this in that server, the leaker, we're assuming, is someone that he was calling OG. He was like, yeah, there's this guy, call me OG. He has money and he's well-armed. He's like this really cool dude. And we played lots of games together on Steam. So Steam is a gaming platform. It's kind of like, it wasn't like
the PlayStation store or like the Xbox game store. He's like, we played Arma 3 together and we played Project Zomboid and we played Halo. And he mentioned all these games they were playing. And that's like how like they're, they're buddies and they just play game together. Right.
So it's a bunch of teenagers, not just Americans, from all over, who are on this little Discord server, about 20 active users or so. And then comes this guy who we now know as Jack. He came in and was sharing all these leaked documents, like, hey, look what's going on, guys, right? That sort of thing. So I knew a handful of people who were involved. So I knew that these guys all played Steam together. I knew some of the games they played. And I, you know...
all that. So I used a few different tools to find, because obviously these people almost all made their accounts private or deleted their accounts basically everywhere after the story broke. So there's a couple sites where like steamid.uk, for example, that has historic scrapes of Steam data. Not just who their current friends are, but who their deleted friends are. Either people who deleted their account went private or removed them from their friends list.
And I looked at some of these people's accounts and I kind of like, okay, here's 400 friends, here's 250 friends, here's 300 friends. And like, who are the intersections? Like who's common friends with all of them? And basically all the people who were common friends with all the players that I knew about were people who were also in the server. So like, okay, well, which of these is the most likely guy?
And there's one guy in particular who kind of stood out. I looked into a little bit more into this guy and I found, actually one of my colleagues over at Bellingcat was looking into his ID and they found that he was on this like arm, like this gun kit selling website. Like he used one of his steam usernames. He also used for selling some kit like scopes and body armor, that sort of stuff.
So we looked more into this guy's Steam list, and we saw, I guess, a few things that matched up. Like all the games that this guy who was on the server told me they played together, Arma 3, Project Zomboid, Counter-Strike, blah, blah, blah. He played all those games. Like there were screenshots he had posted, like there's activity. He played those games. And so then I went through the usernames that he had for Steam, because Steam, people, you can change your username over and over and over and over. And the very first username he had was something that's like very different than the other. It was like Jack something something text.
That kind of sticks out because the other names are like the Excalibur effect, you know, very like dramatic names. And so I Googled this name and there's two results, one scene profile and two was this Flickr album showing a guy and his kid. And this is from 2007. And it was Jack and Jackie was the name of the album, for example.
So I was like, well, Jack is the name on the SIEM ID. But 2007 is pretty old. So at first we thought that this dad was the leaker. But then we looked more and more and we realized that we looked into the mom. And then we found some other family members. And we found photos of a new Air Force Reserve person whose name was Jack, who had just recently entered into the Air Force National Guard.
And he had like some kind of like shit posting adjacent stuff on some of his like his Instagram and his Facebook. And I was like, oh, this makes way more sense. And then we looked a little bit more and we saw that one of the usernames for his Instagram and his Facebook was also something linked back to his Steam account too. So we realized that he just hijacked his dad's Steam account. So the Steam username and the Flickr username were the same.
That's because imagine you're like 12 or 13 and you play your dad's Steam account, but you don't want to make a new one because you already bought games on the account, right? You don't want to make a new account where you lose access to all your Counter-Strike and Bugbee and all that stuff, right? So he just basically took over his dad's account, renamed it, and started using it. But the sites retained the historic name info. So the dad's Flickr account and the Steam account were the same, which led to the family, which led to photos of him. And it linked back to the username he used for his Instagram and his Facebook.
for the same as for his kind of revised, hijacked, same account. And then from there, we realized, oh, he's actually with the intelligence wing of this international guard. And then everything kind of fell together. Wow.
And besides deleting the server, the Thug Shaker Central server off of Discord, did the leaker try to cover his tracks in any other ways? I mean, he made his Steam profile private, but there are services out there, steamid.uk. Big shout out to those guys who run that server. I'm a $9 a month Patreon subscriber, so I got access to their premium stuff to help with this story. So...
Eric, this is a very impressive bit of internet sleuthing. And I know that you have spent a lot of time tracking down not just this leaker, but many other people and incidents on the internet. That's kind of your specialty. But I think the details of this are surprising to a lot of people just because it seems like such a mismatch.
Usually when you have a leak of military intelligence, it's coming from someone with an axe to grind or a point to prove, a whistleblower who has some conscience-driven reason that they want to get this information out there. And it just seems so strange that it emerged in this Discord chat, Thug Shaker Central, where these gamers were—many of whom were literally teenagers—
We're just sort of hanging out, trading insults and memes and whatever else. So help us understand, for people who have not spent as much time in these kinds of Discord servers as you have, what is the culture of a server like Thug Shaker Central that would make someone like this alleged leaker, Jack Teixeira, want to post anything?
classified military documents so i mean it you know again they were just you know like 20 or so kids and they just kind of it's very intimate environment because like it's like voice chat too right you're doing a lot of voice chat you're playing games and all that stuff so it's not like it's like this very like i don't know like sanitized neutral like distant like a lot of times like twitter or like you know whatever is a little bit more distant right you see tweets and you interact with them very publicly this is a lot more intimate you have voice chat things like that
So because of that, people probably felt like they were very close to one another. I mean, like this, the kid who gave the interview to Washington Post said, like, he was my best friend. He felt like he was very tight, very close to him.
I'm not sure if it was reciprocated, but you can kind of get the idea that there was this sense that they were a tight-knit community. So think of it as a more intimate group chat, I guess you could say, rather than a Discord server, which makes it sound like it's a lot more official than it really was. Which kind of builds the trust, I guess. I guess what I'm wondering is, I've been in some gaming group chats before,
I have friends that I know through playing video games. I have never once thought to leak them secret information from my job. You're too old. You're too old to think that, yeah. But sort of zooming out, as a person who has been studying extremism and online mischief for a long time, who has sort of specialized in tracking down hard-to-find people and things on the internet...
Was it surprising at all for you that this new leak that is really turning the United States government upside down, was it surprising to you that it came from such a seemingly strange source?
Not really. I mean, if you think about like, how is this leak going to happen? Because I'm sure this happens. I'm sure that there are people who have classified documents who share them with their buds, like podcast up, right? But most people don't have a 17 year old friend who will then post them onto a different discord. So maybe the people who are doing this right now just have better friends, or at least maybe friends with better filters or sense of judgment. Hmm.
I mean, I, well, yeah, one lesson I'm taking from this is that people are just not as good at doing crimes as they think they are. And that especially when the internet is involved, they just like do not understand how to share things with other people in a secure way. Yeah. It's insane. Yeah. Eric, thank you for coming on hard fork. Sure. Thanks for having me.
Welcome to the new era of PCs, supercharged by Snapdragon X Elite processors. Are you and your team overwhelmed by deadlines and deliverables? Copilot Plus PCs powered by Snapdragon will revolutionize your workflow. Experience best-in-class performance and efficiency with the new powerful NPU and two times the CPU cores, ensuring your team can not only do more, but achieve more. Enjoy groundbreaking multi-day battery life, built-in AI for next-level experiences, and enterprise chip-to-cloud security.
Give your team the power of limitless potential with Snapdragon. To learn more, visit qualcomm.com slash snapdragonhardfork. Hello, this is Yuande Kamalafa from New York Times Cooking, and I'm sitting on a blanket with Melissa Clark. And we're having a picnic using recipes that feature some of our favorite summer produce. Yuande, what'd you bring? So this is a cucumber agua fresca. It's made with fresh cucumbers, ginger, and lime.
How did you get it so green? I kept the cucumber skins on and pureed the entire thing. It's really easy to put together and it's something that you can do in advance. Oh, it is so refreshing. What'd you bring, Melissa?
Well, strawberries are extra delicious this time of year, so I brought my little strawberry almond cakes. Oh, yum. I roast the strawberries before I mix them into the batter. It helps condense the berries' juices and stops them from leaking all over and getting the crumb too soft. Mmm. You get little pockets of concentrated strawberry flavor. That tastes amazing. Oh, thanks. New York Times Cooking has so many easy recipes to fit your summer plans. Find them all at NYTCooking.com. I have sticky strawberry juice all over my fingers.
Casey, it has been several weeks since we have talked about Twitter on this show. We had to set it aside for a minute and sort of let it cool down. Right, it went into timeout, but now it's back, and a lot has been happening.
So I have to confess that a few months ago, I was kind of doubting whether all the predictions of Twitter's doom were actually going to come true. Like people kept saying the app is dying. People are leaving. It's going to break. They're laying off all these people. It still ran and seemed pretty normal.
But now it looks like things are seriously starting to decline. And we don't have time to run through all of the disasters. So I just picked sort of three categories of Twitter chaos that I want to ask you about. Check marks, circles, and clown shit. The three Cs.
So let's start with checkmarks. Okay. So we're talking about the checkmarks, which are officially called verified badges. And Elon Musk, you know, he said that on April 1st, every person who does not pay for Twitter Blue who has one of these checkmarks, you know, members of the media or celebrities or politicians would lose their checkmarks. That date came and went. Yeah. And as of right now, I'm pulling up my Twitter profile and I still...
have a checkmark. Oh, look, I do too. So what is the deal? What is going on with Elon Musk and these checkmarks? Well, look, on one level, checkmarks did play an important function on the old Twitter because they would tell you that somebody was who they said they were. And that helped with impersonation. And then Elon decided that the
best path forward for Twitter as a business was to pivot it away from advertising and towards subscriptions. And the main kind of carrot that he wanted to offer to people in order to get them to subscribe was a verification badge of their own. As you well know, the first rounds of this went very poorly. They let anyone buy a badge for $8. They didn't check to see who anyone was. People impersonated people
You know, the pharmaceutical company Eli Lilly and many other institutions at some point causing wild swings in their stock price. And so it kind of got put on a back burner. And Elon said, let us think this through a little bit more. And so, yeah, as you say, the new deal was going to be, okay, everybody, April 1st, if you're not paying, you're going to lose your badge.
And then that got delayed. And so now he's saying everyone's going to lose their badges on April 20th. That's 420 for all you stoners out there. And we'll see. I mean, you know, Elon has a very chat GPT-like relationship with the truth. You know, like this is somebody who is not drawing on a store of facts all the time, I would say. And so when I hear that he has set a new deadline, no part of me believes that it's real. I think it could happen, but like I don't have any expectation that it will. Right.
What do you think is actually going to happen on 4/20, the new day when all the checkmarks are supposed to disappear from these legacy verified people and organizations? Well, here's what we know. So far, only a very small percentage of Twitter users are paying for Twitter Blue, which includes this badge.
He keeps extending the deadline, I think, because he assumes people are going to fall in line and they're going to say, oh, I've gotten my extension. I guess I'll get that eight bucks together and I'll finally pay. And he's just banking on the fact that these badges are going to become more precious to people. I don't think that's going to happen. So at some point, he's going to have to put his money where his mouth is. He's just going to have to start taking away people's badges and then see if that has the desired effect. I don't think it's going to. But
I can also see why he would want to delay it because after he does that, he doesn't have a lot of moves left, right? I mean, you could imagine moves, like he could start charging people to tweet, right? Something like that. But if we get to that point, it's going to be in a very desperate place. So I could see him delaying this again and again and again. Yeah. My bigger question is like, we know that Elon Musk wanted to shift
Twitter from an ad-based model to a subscription model, right? That is the business problem that he is trying to solve by instituting this $8 a month, like, Twitter blue verification scheme. Does the fact that this program has been so unpopular mean that people just aren't willing to pay for social media at all?
I think people are willing to pay for all sorts of things on social media. It tends not to be these features that were previously offered for free, but TikTok was the highest grossing app last year. And the reason isn't because you have to pay for your badge. So there's a lot of money out there for the social media ecosystem, but
you have to bring real product and design chops to it. You have to offer a stable functioning product. And Elantus hasn't been able to deliver on either of those. Yeah, I mean, to me, it shows the power of anchoring, right? If you start a new social app and you say, you know, this is going to cost $8 a month and here are all the things you get, or, you know, you can give tips to your favorite creators. Like that seems like
a proposition that people might actually go for. But if you take something that used to be free and then start charging for it without any real increase in value, like people are anchored to that free price and it's going to seem not just a stretch for them, but it's going to seem offensive to them. They're like, I used to get this for free. I'm giving you my tweets for free. You're using them to make money. Like I'm not going to,
pay you for the privilege of filling up your platform with my content. Totally. And to me, the exciting thing here is this is a move that is finally making people start to say, you know what? Maybe I should stop giving this guy my tweets for free. All right. Next C on the list circles, Twitter circles.
are a product that I have never used. What are they? So Twitter Circles is very similar to the close friends feature on Instagram. So if you use Instagram, you can create a list of your very close friends. And then if you post to your story, these sort of ephemeral posts that only last for 24 hours, those stories will only be seen by people in your close friends.
Twitter circles launched with the same basic idea that instead of broadcasting your tweet to the entire world, you could broadcast to a subset of the people that you follow. Now, I never use this feature very much myself because most of the people that I would include in my close friends are media people. And if I said something stupid on Twitter to my close friend circle, they would just screenshot it and get me fired. This is a huge relief to me, by the way, because I was paranoid that you did have a close friend circle, uh,
on Twitter and that I wasn't in it. So I was posting detailed reviews of your podcast performance every week to a close group of friends. No, I wasn't using it, but I'll tell you, I have some gay friends in particular who were using circles to post, let's say some spicier, more adult oriented content. And so that made what happened on Twitter circles this week,
something of a catastrophe. Okay, what happened? So what happened was, as I said, these posterior circles are supposed to be limited to a group of people that you've put on an allow list, okay?
But then some poor souls browsing the Twitter feed started to see posts on the For You page. The For You page is the algorithmically ranked feed of posts that Twitter might think you would enjoy for any number of reasons, right? It's ranked, but it could potentially be seen by anyone. And so essentially, you had private posts that were being not just made public, but
but recommended to the public. So the best analog I can offer here is imagine that you posted something really heartfelt or embarrassing or vulnerable on your close friend's Instagram story and then a stranger opened up the Explore page on Instagram and saw it promoted at the top of their feed. That's what just happened on Twitter circles. And
And in a way, it's like funny to me that this hasn't gotten more discussion because this is a truly catastrophic privacy breach. Right. I mean, it's sort of like if your DMs were just showing up on people's Twitter timelines. It's like that level of a privacy scandal. Yeah, absolutely. You know, Elon did an interview this week where he boasted that a lot of folks said Twitter would be dead by now, but it's not dead because like his standard for his Twitter alive, I think, is like, can you still log into the website? Yeah.
But this is a great example of how when you get rid of a huge number of people that worked on your technical infrastructure, weird stuff is just going to start breaking in ways that can have quite a terrible effect on your user base. Yeah, I mean, I've been wondering whether there will be some kind of mass DM leak that
that stems from all these engineers getting laid off. That hasn't happened, but this Twitter circles thing does seem to be pretty close. Like these are supposed to be private, but all of a sudden started showing up on people's For You feeds. Oh yeah, and when the layoff started, Twitter engineers started messaging me saying like, don't use DMs anymore. Like they're just not a safe product. So I don't use Twitter DMs anymore for anything remotely sensitive. But yeah, watch out, man.
Wow. So do you have any idea of what went on behind the scenes to make these Twitter circles posts start showing up for people they weren't supposed to be seen by? To my knowledge, they have said nothing about it, which does not exactly inspire confidence. So put your spicy tweets somewhere else because they are not safe in Twitter circles. Yeah. OnlyFans is now accepting spicy private messages.
Subscription only posts. When are you starting your OnlyFans? Well, we'll see how Substack goes. Okay, that brings me to my third kind of catch-all category for what I am calling clown shit. This would include things that Elon Musk has done in the past couple of weeks that just seem like he's just kind of playing for laughs or not behaving in a serious way. He...
briefly changed the Twitter logo to the Doge image, the Shiba Inu dog that is also the face of Dogecoin. That lasted for a few days. It was not on April Fool's Day. No, I believe it launched on April 3rd, and everyone was joking like Twitter tried to do an April Fool's joke, but it took like an extra two days to launch. Yeah.
Which seems plausible to me. Yeah, totally. He also painted the sign on the Twitter building in downtown San Francisco, or he had it painted so that it just now reads Titter. I truly don't believe that even an eighth grader would find this stuff funny.
I can't imagine anyone is observing any of this and thinking, this is so funny and cool. Right. This is the man that I want to entrust my company's advertising budget to. He also changed his display name on Twitter for a day or two recently to Harry Balls. Harry spelled H-A-R-R-Y and Balls spelled B-A-L-L-S.
B-O-L-Z. This man is not well. Can we just say the man is not well? Yeah, what is going on? Look, I think that driving a social network can actually drive you crazy. It's very difficult. You're running a platform...
that is hosting this huge portion of the world's speech. You're getting criticism from absolutely every corner, and that's hard enough even if you are a very good CEO with kind of a good plan for where you're going. I think if you come into that with no background in the subject matter, and you've committed $44 billion, and you are losing money every month, and none of your ideas are working,
Oh, and also you're working on a handful of other giant companies that have their own problems. That is just not a recipe for success, right? And so many people when Elon took over said, well, look, this guy's going to be different. Okay, he's special. He builds rockets. A social network? Yeah, I think he can handle that. And then you fast forward to April of 2023. Guess what, honey? It didn't work. He's like, give me the rockets back. That's right.
Right. This isn't rocket science. It's actually harder. It is hard. But like, is there one person who wouldn't be relieved if Elon Musk says, I am stepping away from Twitter and I'm just going to work on rockets full time? Like everyone would be overjoyed. I think even his fans would be overjoyed because he's been getting blowback from some people who previously have supported him. People like Paul Graham, who is a big Silicon Valley investor. He's a co-founder of Y Combinator.
And he's been criticizing Elon recently for basically just taking his eye off the ball, you know, like wasting his time running Twitter instead of working on SpaceX or Tesla.
So even the people who are sort of predisposed to support him are not fans of what he's doing at Twitter. I want to ask you about one more thing that I would put in the clown shit category, which is what is going on between Twitter and Substack? Well, so, you know, at this disclosure, I published my newsletter, Platformer, on Substack, and they've helped me in various ways over the years, and I have a note on my website about that. But...
decided in the midst of all this chaos to... And Substack, for the people who may not know what it is, it is a newsletter platform. You can subscribe to newsletters, either free or paid. Substack is just kind of the infrastructure that powers that and allows people to pay you to read what you have to write. Yeah, that's right. And they saw everything going on at Twitter and they said, well, gee, it seems like this thing is having some trouble. Maybe we could build
something similar of our own. And so they built this product called Notes. It launched this week. It lets you post text and links and pictures, and you can reshare those. They call that restacking. And when Elon got word of this, he really overreacted. And for a time, he blocked users a
on Twitter from engaging with tweets that had links to sub stacks in them. If you click the link, it would say this website is spammy or dangerous, which is true in the instance of platformers as a spammy and dangerous newsletters. It's better unfair. Yeah, it's unfair.
So they marked Substack links as, quote, potentially spammy or unsafe. They also blocked searches for Substack. So if you search the word Substack, it would just show you results for the word newsletter. So it's like, oh, you're looking for a Substack? Well, we're not going to show you that, but we will show you newsletters. So when he got asked about this, Musk said that Substack was, quote, trying to download a massive portion of the Twitter database to bootstrap their Twitter clone, which
And I went over to Hacker News to see what they were saying about this comment, which I found completely inscrutable. And the Hacker News people, this is a community of engineers and some of the most delightful nerds you'll ever meet. And the Hacker News people were just laughing their heads off because they were like, what does this even mean? It's like, you can't download it. Like, it just didn't make any sense whatsoever. And so, you know,
as usual with him, there's not a relationship with the truth there. It's just kind of like, you know, spewing words and it was all nonsense. I mean, what I found most interesting about this beef between Elon Musk and Substack is that one of the things that animated him when he took over Twitter was that he thought that the previous management of Twitter was sort of
throttling or shadow banning sources of news that they didn't like. That's right. He was very upset about what happened with Hunter Biden and that story during the 2020 election where Twitter blocked people from sharing that New York Post story about Hunter Biden. Right. And then...
He Hunter Biden'd Substack. He made it so that you could not share Substack links or comment on them or like them or search for them. He disappeared Substack from the platform. And Substack is where a lot of the people that he is ideologically allied with, these sort of Twitter files, journalists and other opponents and skeptics of the mainstream media are going. He was basically cutting off his friends and allies from one of their main sources of distribution.
It truly is the funniest possible outcome to this story. And I just think the more serious point is this is not a principled man. This is someone who talked about nothing except his speech principles and his desire for bipartisan dialogue and let's move to the center. And he spoke about it in the most grandiose terms.
And the minute he felt the slightest bit of discomfort, all those principles went out the window, and he said, I am going to punish my enemies. And
Look, some people are very upset about this. I can't get upset about it because every day people learn what I truly have been saying all along, which is like the emperor has no clothes here. And you can just go on Twitter every day now and see it for yourself. But one of the other things that you have said is that this sort of chaos at Twitter is going to result in people moving. Yeah. So far, that hasn't.
hasn't really happened. There has been no mass exodus of Twitter users to another platform. You're very active on Mastodon now. I think it's safe to say that has not really threatened Twitter in a massive way. Now, people on Twitter are talking about Blue Sky, which is this sort of dehumanizing
decentralized alternative to Twitter that was started by Jack Dorsey. But none of these alternatives to Twitter have really gained momentum. So what is it going to take for something to pop up and replace Twitter? Or is it just kind of like the entire concept of Twitter in our
media ecosystem sort of dies off and is replaced by, I don't know, group chats and Instagram accounts and whatever else. So there's two answers to that. There's the product answer and there's the NPR answer. I want to talk about the product answer first, right? So you just named some products that compete with Twitter in various ways. And you're right. None of them has supplanted Twitter and the popular imagination. But again, let's remember, Elon has only owned this website since October. We are six months out of
from when he took over. We first started talking about this on Hard Fork. The reason that these products are now being built is because there is an appetite for them. They are growing. And frankly, there are just now so many of them that even if one of them hasn't entered that period of hockey stick growth quite yet,
I do think it is probably going to happen for someone. So you're going to have to give it some time. Someone is probably going to have to invent some fun, new little mechanic. There's going to have to be some weird, silly new toy that just kind of captures our imagination. But I do think it's going to happen. So first answer is these products are still getting built, but they are getting built. And that was not true before Elon Musk took over, right? Nobody smelled blood in the water until the past few months. At the same time, Kevin,
Some important people actually are starting to leave. And I want to deliver one of our signature hard fork tributes this week to a little media organization called National Public Radio. You know these guys? Yeah, NPR. I've heard of them. NPR, they were just minding their own business. They were just tweeting links to their stories.
And then one day they wake up and they've been slapped with this label, state-affiliated media. And state-affiliated media was a label that the old Twitter created to identify essentially propaganda. Right. It was like a little label that went under the profile of like essentially foreign propaganda outlets just to let you know that the tweets that you were seeing from this account have –
state affiliation. Yeah, it's just trying to give you a little bit more information so you know where the account is coming from. They allowed them to post, but they wanted you to know that. Well, all of a sudden, NPR, which does receive some minor amount of government funding, but is completely editorially independent of the United States government, they now have this label as well. So it
said, this is propaganda. That was the message that I believe the people who added this label to NPR's account wanted you to receive. And messages are exchanged. And eventually, the state-affiliated media label gets downgraded to the government-funded media label, even though it gets less than 1% of its funding from the Corporation for Public Broadcasting.
So the entire point of this was not to serve users of Twitter in any way. It was just to slag NPR and to try to make it look less credible. Right, it was a dunk. It was a dunk. And on Wednesday, Steve Inskeep tweeted a picture of John Lansing, who is the CEO of NPR, at a Morning Edition meeting, Morning Edition, one of the flagship shows of NPR, and said that NPR will de-emphasize Twitter.
He said that aside from the misleading label, Twitter isn't used by most Americans, drives little traffic to NPR, and no longer has the public service relevance that it once had. And to that, I just want to say,
Slay. Well said. This is what I have been saying, and pretty soon a lot more people are going to get it too. You think media organizations are going to leave Twitter? This is what I have been waiting for, is for somebody to wake up and say, I do not want to come to a platform every single day and get punched in the face. You know,
We know what I've been thinking about lately, Kevin. Do you remember during the Trump campaigns when there would be these rallies and in the center of the rally, there would be a pit for the media and a signature moment of every rally would be Trump pointing to the people in the media so that everyone at the rally could say, boo, we don't like the press.
That is what Twitter has become. It is the press pit where a bunch of people are standing around you in a circle jeering. Adding this state-funded media badge was one of those steps. But I'm barely joking when I say that I think eventually every reporter who is still on the service will have a clown badge next to their username. And you just have to decide if you still want to be there when that happens. Hmm.
So you are still on Twitter. Well, in the way that I started with. So in December, I was like, I got to change the way that I'm using this website. And so I basically started using it as an RSS feed. Whenever I have a sort of Twitter-like thought, I put it on Mastodon. And when I have some new piece of journalism to share, a newsletter, a podcast, I put it on Mastodon and then I repost it to Twitter. I'm not in there having fun. I'm...
really not replying to people. I'm reading it because it feels important to know what is being said on Twitter, and it's important to know how Twitter is falling apart. Social networks are my beat. I have to pay close attention to it, but I am not really doing free labor for Elon Musk anymore. Fascinating. But you still are. Let's talk about that. Well, I want the clown badge. That would be fun for me.
Okay, so Twitter is in Clown Town right now. But next, I want to talk about a different town. One that is populated entirely by AI. Uh-oh! We'll be right back. ♪
Indeed believes that better work begins with better hiring, and better hiring begins with finding candidates with the right skills. But if you're like most hiring managers, those skills are harder to find than you thought. Using AI and its matching technology, Indeed is helping employers hire faster and more confidently. By featuring job seeker skills, employers can use Indeed's AI matching technology to pinpoint candidates perfect for the role. That leaves hiring managers more time to focus on what's really important, connecting with candidates at a human level.
Learn more at indeed.com slash hire.
Christine, have you ever bought something and thought, wow, this product actually made my life better? Totally. And usually I find those products through Wirecutter. Yeah, but you work here. We both do. We're the hosts of The Wirecutter Show from The New York Times. It's our job to research, test, and vet products and then recommend our favorites. We'll talk to members of our team of 140 journalists to bring you the very best product recommendations in every category that will actually make your life better. The Wirecutter Show, available wherever you get podcasts.
Casey, did you watch Westworld, the HBO show? Yeah, I watched the first few seasons, but I gave up as it became increasingly incomprehensible. Okay, so I also watched like one season and I think it fell off because I just didn't get it anymore, which I attributed to me being dumb, but I think maybe the show just also got worse. Well, no, here's how it became one of those shows where it was like, at first it was very exciting to learn that that person you thought was a human was actually a robot, but then that was just the entire show. It was like, well, guess what other human is a robot?
every member of the cast. Right. So for people who haven't watched Westworld, the premise is there's this like theme park where there are humanoid robots that you can go and interact with and they'll talk to you. They're perfectly lifelike. Perfectly lifelike. But when the guests leave, the robots' memories all reset. And so the next time you go, you can just kind of have like brand new interactions with them.
Except what Westworld asks is, what if they didn't actually forget? Exactly. So I want to talk about this next story because I think it is actually moving us in quite an interesting direction that is a lot like Westworld. So a paper came out last week by researchers from Stanford and Google, and the title of the paper is called Generative Agents, Interactive Simulacra of Human Behavior. Another one of these catchy clickbait titles from the academics. Yeah.
And I want to talk about this paper because it's really fascinating, and I think it actually points to the next phase of what's going to happen in AI.
We've been talking a lot on the show about these large language models, these chat bots, things like ChatGPT, and how you can actually use them as a kind of persona. Like you can say, you are my tutor, or you are my speech writer, or you are my editor, or you are my friend. Talk to me as if you are that thing.
And what these researchers wondered is, could you create a city, like a virtual simulated environment, and fill it with AI-generated characters? A simulated city or SimCity, if you will. Right. Basically, could you make The Sims, but The Sims aren't coming up with things on the fly. So The Sims is this, you know, very popular computer game. But in The Sims, like all of the interactions between characters are scripted and programmed. Right.
So what these researchers were wondering is, could you do The Sims with no scripting? If you just took a set of AI-generated personas, which they're calling generative agents, and placed them into a virtual environment, like a city, and just pressed play...
What would they do? I have to say, very interesting question. So the way this works, they created 25 characters. They put them in a virtual environment that they called Smallville, which is kind of like a kind of Animal Crossing-style video game.
And they gave all 25 of these characters short personality descriptions. So one example was a character named John Lin, and the thing that the researchers told the AI, and in this case, they did this using the ChatGPT API. So all this is happening with the help of ChatGPT. So the thing that they put in as the sort of description of this character was, John Lin is a pharmacy shopkeeper at the Willow Market and Pharmacy who loves to help people.
He is always looking for ways to make the process of getting medication easier for his customers. John Lin is living with his wife, May Lin, who is a college professor, and son, Eddie Lin, who is a student studying music theory. John Lin loves his family very much. This is very sweet. So they build these sort of like pre-programmed personalities for these characters, and then they set them loose. They just say like, you know, we're going to observe you for two days, and
And we're just going to see what these characters will do. And within the context of this game, like, are there objectives? Are these characters moving throughout the world? Are they just sort of bumping into each other on the street and striking up conversations? Or what is happening? Yeah, so the way it works, the researchers who are controlling this experiment can sort of prompt the characters. They can say, you are going to the store to get some food.
Then the language model sort of fills in what happens next. Or you're a character, you're bumping into another character in the park, and you're going to have a conversation. What does that conversation say? And then the language model sort of fills in the rest of the conversation and any subsequent actions. Got it. So...
I can actually show you what this looked like. Let's see this. I'll share my screen here. All right. So this is a sort of replay of the Smallville. So you can see it's sort of like a blocky, like pixelated map. Yep. With some various rooms. And then you can scroll down here and here is a list of characters, 25 characters. And they're these sort of like very cute little sprites.
Right. And so you can click on any of the characters and see a log of all of their actions. So here's a character named Ryan Park and...
Ryan Park's current action is sleeping, and he is in his apartment in the bedroom. And that's goals. Cheers to you, Ryan Park. And here is another character named Sam Moore. Sam is taking a walk around Johnson Park. He's stopping to take pictures. Oh, and he's having a conversation with another character named Adam Smith. And all of this is being generated by
by the language model. And along with these characters, the researchers actually developed something really interesting that is sort of like a memory, like a kind of virtual memory for these characters. Because one of the problems with language models, one of the limitations,
is that they don't remember things. So you can have a conversation with ChatGPT, you can tell it some things, and then if you come back a different day and start a new thread, like it's not going to remember what you talked about. You have to kind of re-instruct it to be a therapist or a teacher or a friend or a negotiating counterparty or whatever.
So they sort of get around this with something that they call the memory stream, which is basically a natural language list, like a text list of what a given character has been doing and experiencing. Any conversation they've had gets sort of stored in this memory file. And then periodically, these memories get kind of synthesized into inferences and reflections, sort of summarized. So you basically have...
a system where each of these 25 characters
can refer back to a file that contains sort of a synthesis and a summary of everything that has happened to them in this simulation. And they can refer to that in different contexts. So, you know, if one character is walking through the park and sees two other characters talking, they can like infer that those people must be friends and must know and like each other. And then they can use that information later on in the simulation. So it's sort of a neat technical trick that simulates a kind of memory. Okay.
Okay. So these researchers, they let this experiment run for two days and they just observe what these characters do prompted by this large language model. What they found was really interesting. So they did a control group experiment where after this two-day period was up,
The researchers asked the generative agents a series of questions to kind of assess their self-knowledge, their ability to make plans and remember things. They also gave that test to a group of humans, 25 humans who had access to all of the same information about these characters.
And they basically gave them a replay of what the characters had been doing for these two days and all of their memories. And so the humans answered those questions. And then the researchers had these other humans, these evaluators, go in and rate how believable they thought those answers were.
And the generative agents were evaluated as being more believable than the humans. So that's bad news for humans, so go on. So that was one finding. The other very strange finding that they had was that there were these emergent behaviors that they did not expect and did not intend to happen between these generative agents. Okay, if you tell me that one of these things started asking if it was part of a simulation, I'm going to need a drink. Okay.
No, they started leaking military documents. So one of the emergent behaviors was that these generative agents gossip. What? They shared information with each other without prompting. You're telling me that these Sims are spilling the tea on each other? These Sims are spilling the tea. So in one interaction, the researchers told one character, this character named Sam Moore, that he was going to run for mayor of Smallville.
Sam Moore then shares this information with the rest of the town and starts a mayoral campaign.
And the other agents start talking about this without interacting with Sam Moore. They're just like, they have learned this information. I mean, this is admittedly big news in a town where, for the most part, people are just sleeping and talking on a park bench. Like, I could see why this would sort of be the buzz in Smallville. But not only are they talking about the fact that Sam Moore is running for mayor, but they're sharing opinions of him. So one agent at one point says...
I'm still weighing my options, but I've been discussing the election with Sam Moore. What are your thoughts on him? And another agent says, to be honest, I don't like Sam Moore. I think he's out of touch with the community and doesn't have our best interests at heart. Well, go off, King.
And not only are they sort of gossiping about each other, but they're also coordinating and planning together. So in one case, the researchers instructed an agent named Isabella to throw a Valentine's Day party. That was the instruction that they gave this character. And this character, Isabella, starts spreading the word. And by the end of the simulation, 12 characters know about this party.
And how many of you do you think showed up? I mean, in the real world, a lot of people would flake, but I'm hoping that these AI agents are more respectful. So I want to hope that all 12 showed up. Well, you are wrong because these are believable entities. And much like humans, many of them flaked. Seven characters flaked on the party. Three of them said they had other plans and the other four just didn't show. It's like setting up a Grider day. This is horrible. Yeah.
but five characters actually did show up at this party. And, uh, this was not something that the researchers had predicted. So, um, why are we talking about this? You may be asking. Well, I assume that each week you show up and try to terrify me with some new glimpse at our coming dystopia. So, um, I,
I do think this has dystopian implications, but I think the most obvious application for this is something like gaming. Absolutely. So in games, a lot of them, especially role playing games or open world games, there are these things called NPCs or non-playable characters. And they're the kind of characters that you, you know, come across and they tell you some line or give you some hint or like, how would you describe an NPC? Yeah. An NPC is essentially an
in the game that exists to give you an objective or to advance you along a quest, but they are scripted to make you feel like you are interacting with something sentient. And of course, you know it's not
But the question that this poses is, well, what if word got around about what you were doing in this virtual world? What if that character could ask you about that or maybe change their behavior? This sort of thing does already happen, but all of those interactions are scripted, right? What you're sort of suggesting here is that
We might not have to script these things in the future. They can just happen. Right. They could just act as the language model tells them to act. So one example that was sort of a joke, but I think someone said in response to this paper that you could be playing a game now and you could kill an NPC in the game and you could return back to that same spot later and all the other NPCs could be holding a funeral.
And, like, that's not something that the game designers would have to necessarily script. I did talk to the lead author of this paper, who's a Stanford researcher named Jun Sung Park. And I sort of said, like, well, beyond games, like, how could this be useful? And he said that one of the ways that this could be used was...
was to simulate behaviors in online groups. So for example, if you were gonna start a new subreddit devoted to like celebrity gossip,
And you wanted to see like, okay, how would a set of rules given to users of this subreddit, you know, encourage them to react? Would they become rebellious? Would the thing sort of devolve into a toxic cesspool? Like, how could you basically simulate that without actually having to start the subreddit? And this experiment suggests that you could actually just start
of simulated subreddit and make some simulated users and see what happens. Totally. You know, another thought that comes to mind is I think we both believe that AI companions are going to be a really big deal. There's going to be some kind of virtual agent that you're talking to, whether it's on your phone or your desktop. And in order for that to work, they need to have a memory. They need to remember what kind of day you were having yesterday. They need to know the names of your friends and family and other people who are important to you.
What I think is interesting about this is it suggests, what if you actually have multiple different companions for different purposes, right? You're going to have your personal trainer and your therapist. And of course, at first, none of these are going to be as good as humans, but they probably all get better over time. And what if they talk to each other and share information?
What if you have a virtual doctor who remembers your entire health history and doesn't make you write down your birthday every time you go to the office, right? So I think there is a huge amount of potential here, and that is a very fun glimpse into somebody already running the experiment. Well, so yeah, so this is the experiment that got a lot of attention this week, but there's also this sort of related thing that is happening in the AI world right now, which is that a lot of people are trying to make these large language models into agents.
So turning them from chatbots into things that can act autonomously without human intervention. We talked a fair bit about like the need for AI vocabulary. And I'm telling you, agent is just one of those words that you are going to start hearing absolutely all the time. Yes. So in the last couple of weeks, there have been a number of projects that have tried to basically counteract
cut out the human. So right now, if you go use ChatGPT, like you prompt it, you type something in, it types something back, you type something in, it types something back. It's like a kind of ping pong. And what these experiments with AI agents are asking is like, what if the AI model could essentially talk to itself?
and come up with plans and execute those plans and then generate AI agents to carry out different parts of this task. We are really just determined to wipe out humanity, aren't we? Well, some people literally are. So right now there's a lot of research and energy in the AI world around turning AI
these large language models into what are called agents. One of them is called AutoGPT, and AutoGPT is a tool that someone hacked together that allows you to, instead of just typing to a chatbot and it types back, you actually give it a task and the AI can then talk to itself and recursively execute that task or try to execute that task.
So you could tell, theoretically, this auto GPT, like start a dropshipping business. And you could then sort of talk to itself and say like, okay, for that, I'm going to need someone who's an expert in shipping. I'm going to need someone who can talk to the big e-commerce companies. I'm going to need someone who can set up, you know, maybe a Stripe.
account. I'm going to need someone to code up a website. And also I'm going to need someone who can go on Twitter and advertise this business. Now, just to be clear, there's no evidence yet that AutoGPT can actually do this stuff well. It's still not very good. And there are still some
pretty big technical hurdles standing in the way of these agents being able to actually do things autonomously. But that's the direction that some people are trying to push these AI systems. Yeah, it's like when you're fighting a boss in a video game and they're like, and you think you're finally whittling them down and then they just summon a bunch of other demons to eat your face. That's what we're building. Right. And the examples of this auto GPT stuff that have been floating around are mostly just kind of people trying to do these business tasks.
But there is one project called Chaos GPT where someone basically used auto GPT and gave it the explicit instruction to try to destroy humanity. Well, the world is still intact as we record. So what happened? So, yeah, this has not worked, as you can see by the fact that we still exist. I mean, my world was destroyed when Taylor Swift and Joe Alwyn broke up. But other than that...
But there is a mode on AutoGPT called continuous mode, which basically allows it to, in theory, run forever, just doing these kind of like recursive task loops. And when you turn on continuous mode, it does actually pop up a little warning saying like, you might not want to do this because it is, quote, potentially dangerous and may cause your AI to run forever or carry out actions you would not usually authorize.
But the mode exists, people are using it, it's out there already. I've talked to some AI safety researchers in the last few days who are very concerned about this shift from thinking of these AI language models as tools to using them to create autonomous agents.
And what they'll say is like, look, these experiments, these Smallville, these generative agents that are just going to parks and talking to each other and planning Valentine's Day parties, like these are not dangerous because, you know, they're contained in this virtual environment and they're pretty safe in the things that they are talking about.
But you actually don't want AI agents operating without human oversight. You don't want them carrying out their own goals. You don't want them doing long-term planning and coordinating with each other. That's just not a capability that you want to build, right? Yes. And while I'm glad that some of this research is being done in the academic realm, the fact
that these kind of experiments are already being run does make me hope that our government is paying attention and is pretty soon going to start to have some ideas about whether and how all this ought to be regulated. Yeah, I mean, I will say, like, I'm pretty skeptical of regulation in this space, at least the proposals that I've seen so far. But when I saw Chaos GPT, which stated that its explicit goal was to destroy the world, I was like, yeah, that should probably be illegal. Yeah, I mean, there's... Yeah, um...
And this is the thing, like all the scenarios of AI doom out there, like imply that the robots are going to have to like become power seeking and, you know, greedy for some resource that it will kill everyone to get. But like, actually, it may be that like some troll on the internet is just going to like give it an instruction to go destroy humanity. And as soon as that's done, you know what the first thing they'll do is? They'll go on Discord and say, hey, look at this. Guess I won. Right. So.
The last concern that I heard from AI safety researchers about these AI generative agents is that these things that start off as being very benign and not all that capable tend to
to get more powerful over time. Like the Hard Fork podcast. Exactly. And so you can imagine hooking some of these generative agents up to physical robots. Like, that is a thing that can be done. Like Megan? Yes. You could take one of these generative AI personas and tell it to control a hardware robot. And like, it could do that. Oh, you know what? My dream of a sassy Roomba has never been closer to reality. Yeah.
So, I really just think this is an area to watch and that I hope we'll come back to on this show because this use of AI to create these autonomous agents, I think, poses some really interesting opportunities. I do think there are applications for this beyond gaming. And I think it's an area where there is the potential for a lot of harm and mischief. So, yeah.
Stay tuned. We will be giving you more updates on the AI agents. In the meantime, I think I want to go check out Smallville, actually. I want to see what the buzz is. Are they talking about us? Are they listening to the show? I don't think so. But, you know, I'm going to check out another character. Maybe we'll find something here. Latoya Williams is currently taking a yoga class, doing standing poses. Everyone in this town is having a more relaxing day than I have. I'm jealous. Let's go to Smallville. All right, let's do it. All right.
BP added more than $130 billion to the U.S. economy over the past two years by making investments from coast to coast. Investments like acquiring America's largest biogas producer, Arkea Energy, and starting up new infrastructure in the Gulf of Mexico. It's and, not or. See what doing both means for energy nationwide at bp.com slash investing in America.
Hard Fork is produced by Davis Land and Rachel Cohn. We're edited by Jen Poyant. This episode was fact-checked by Caitlin Love. Today's show was engineered by Alyssa Moxley. Original music by Dan Powell, Alicia Baitup, Marion Lozano, Rowan Nemisto, and Sophia Landman. Special thanks to Paula Schumann, Nelga Logli, Kate Lepresti, Huiwing Tam, and Jeffrey Miranda. You can email us, as always, at hardfork at nytimes.com.
But please, no top secret documents. Save those for your Discord server.
