메타 AI 조직 내홍, 틸의 비밀 모임, 그리고 트럼프 사면 타진

Comment Loader Save Story Save this story Comment Loader Save Story Save this story This week on Uncanny Valley , our hosts discuss the meltdown that has been recently unfolding at Meta and what it says about the company’s relentless ambitions in the AI race. They also dive into the leaked messages and names of an invite-only group cofounded by billionaire tech founder Peter Thiel, and how Sam Bankman-Fried is now actively seeking a pardon from the Trump administration. Plus, they share their impressions on SpaceX acquiring Cursor and the latest on the negotiations between Anthropic and the government. Articles mentioned in this episode: Meta CTO Andrew Bosworth Admits the Company’s AI Reorg Was ‘Atrocious’ ‘Tell Him He’s a Piece of Shit’: Meta’s New AI Unit Is a Total Mess Leak Exposes Members of Peter Thiel’s Secretive ‘Dialog’ Society Anthropic Is Still at Odds With the White House Over Claude Fable 5 You can follow Brian Barrett on Bluesky at @brbarrett and Zoë Schiffer on Bluesky at @zoeschiffer . Write to us at [email protected] . How to Listen You can always listen to this week's podcast through the audio player on this page, but if you want to subscribe for free to get every episode, here's how: If you're on an iPhone or iPad, open the app called Podcasts, or just tap this link . You can also download an app like Overcast or Pocket Casts and search for “uncanny valley.” We’re on Spotify too. Transcript Note: This is an automated transcript, which may contain errors. Brian Barrett: Hey, this is Brian. Before we start, two quick things. If you've been enjoying listening to the show, we would appreciate it if you took a second to rate it in your podcast app of choice. It really helps us reach more people. And second, if you have any questions related to tech, privacy, or politics that you would like me, Zoë, and Leah to take on, now is the time to submit them to [email protected] . It doesn't matter how big or how small, we want to hear from you and get you answers. OK, on to the show. Zoë Schiffer: Welcome to WIRED's Uncanny Valley . I'm Zoë Schiffer, director of business and industry. Brian Barrett: I'm Brian Barrett, executive editor. Zoë Schiffer: Today on the show, we're talking about the dysfunction in Meta's newly formed AI unit and why it's been driving employee morale, which was already very, very low , even further into the ground. Brian Barrett: We'll also break down the recent online leak that shed light on Peter Thiel's invite-only group, Dialog , more than 200 names of high profile people in government, tech, academia, beyond are listed in the documents as members and guests of this secretive society, not to mention a look at what they talk about behind closed doors. Zoë Schiffer: And it's a busy week for controversial figures because we'll also get into how the former cryptocurrency founder and now convicted felon , Sam Bankman-Fried , is not only trying to make his case to get a pardon from the Trump administration , but planning a potential comeback. Brian Barrett: And a bit later in the show, we'll talk about SpaceX's acquisition of Cursor and the latest on Anthropic's efforts with the Trump administration to get their latest models back online. Zoë Schiffer: Brian, I think we have to get started with the complete mess and mayhem that is going on inside Meta because the company— Brian Barrett: It's bad. Zoë Schiffer: It's bad. It's so bad. People are not happy. Just to set the scene, the company has been laser focused on trying to catch up and maybe win the AI race . It has poured a ton of money into creating a new AI unit, Meta Superintelligence Labs, and they've invested a ton in AI infrastructure, they've released new models—that's been a little bit bumpy—but more recently, the company has reshuffled its staff in favor of prioritizing AI. During the company's most recent round of layoffs last month, about 7,000 people were transferred to work on teams focused on AI. One of these teams was Meta's applied AI engineering unit, which basically supports the work of the people who are in Meta Superintelligence Labs, which is the super fancy AI lab. And perhaps it's an understatement to say things have not gone smoothly at all. Brian Barrett: Just to set the scene too, so you work at Meta, about 8,000 employees were let go as part of that round, right? Zoë Schiffer: Yes, exactly. 10 percent of the company. Brian Barrett: And then another 7,000 scattered to various AI units, including this one Meta applied AI unit, which people hate. They have been—in dramatic fashion too. A source that we talked to called it "the Gulag." Zoë Schiffer: They did say that. Brian Barrett: Which seems a bit much. But last week we learned during an employee-only meeting for the Meta applied AI unit , someone interrupted the call to say that they were "being the company's (beep)." The same person then asked people leading the call to write to a specific Meta AI executive and, "Tell him he's a piece of (beep)." We have both heard this recording, Zoë. Zoë Schiffer: Yes, we have, many times. Brian Barrett: Yes. And as much as I enjoy the interruption, I even more enjoyed the deafening silence immediately after. Zoë Schiffer: Yeah, that was rough, and there's two things to say about this. One is that one of the reasons that people are really upset is that the work they are being asked to do they perceive as being very menial. Meta basically took a bunch of engineers who, at least in their telling to us, were working on interesting, exciting projects, and said, "Hey, you now work in this other team, and your projects are basically solving problems on behalf of an AI. If an AI can't do something, you help it do it." You're doing what essentially sounds like post-training to fine-tune a model and improve it for specific purposes. Employees said things like, "It's not like this work is difficult; in fact, it is that the work is very not challenging. It's chill, but suddenly I have no purpose in life. It feels like I'm just given these random tasks. I don't have agency anymore." And the other thing about it, which you and I have talked about, is employees didn't have a choice about joining this team. Brian Barrett: It's people who had joined Meta because in some cases I'm sure honestly believed in the mission of social media apps that can connect billions of people, whether you or I believe that that higher mission still holds at Meta, people presumably were there, and now they are sort of training the machine. The further context here too, Zoë, which we've talked about a lot but is worth saying, this comes on top of Meta saying, "Hey, by the way, we're going to monitor your laptop usage — Zoë Schiffer: Right. Brian Barrett: —and track what you do also in order to train AI." So if you're a human working at Meta, it suddenly feels like everything you do is just in service of the machine instead of in service of something bigger. Zoë Schiffer: Yeah, and people are frustrated about this for a number of reasons, but one of the reasons that this is annoying to some people at least is that Meta's actually doing remarkably well as a company, it's had record-breaking or near record-breaking quarters, but it's not having business success necessarily because of Meta's AI projects. And it's not to say the ads product doesn't have AI in it, but there are other parts of the business that have been around for a lot longer that are actually doing exceedingly well, and now these employees are saying, "Hey, we actually were doing our jobs pretty well as seen by the fact that the company is printing money, and now you're asking us to go do this other thing that is actually causing us to somewhat lose a bunch of money and force yo