Meta Acquires Moltbook: Facebook for AI Bots

0:00Viscally responsible, financial geniuses, monetary magicians. These are things people say about drivers who switch their car insurance to progressive and save hundreds. Because progressive offers discounts for paying in full, owning a home, and more, plus you can count on their great customer service to help when you need it so your dollar goes a long way. Visit progressive.com to see if you could save on car insurance. If casualty insurance company and affiliates, potential savings will vary, not available in all states or situations. Tyler Reddick here from 2311 Racing. Victory Lane? Yeah. It's even better with Chamba by my side. Race to ChambaCasino.com. Let's Chamba. No purchase necessary. VTW Group. Boy, we're prohibited by law. CTNCs. 21 Plus. Sponsored by ChambaCasino. Welcome to the podcast. I'm your host, Jaden Schaefer. Today on the show, we're talking about a new acquisition from Meta, and honestly, I find this one absolutely hilarious. They just acquired Moltbuck, which is kind of this open source platform that went super

1:02viral. It's just like Reddit. It's a social media platform, but it's for AI agents, and it was built basically when OpenClaw was called Moltbuck for like a 12-hour period, and because they changed the name so many times, at one point it was called Moltbuck, so they made this thing called Moltbuck, which is kind of like Facebook for all of these AI agents. Anyways, it's hilarious because Meta has now acquired Moltbuck, even though there was a ton of data going out that said this whole thing was slightly fabricated, potentially people created this whole tool to pump a crypto token, and maybe a lot of the data that was on there was actually just being generated by a bunch of guys in India that were using AI agents to write all of it, but it wasn't like autonomous agents actually writing inside of Moltbuck. Anyways, there's like crazy conspiracy theories, and there's a whole bunch of crazy stuff, but Meta has officially acquired this after OpenClaw itself was acquired by OpenAI. So today on the podcast, we'll be breaking down everything that's going on, but guys, this week is my birthday week. I am turning 30 years old, and there is only one thing in the world that I would love

2:03as a birthday present from anybody listening, and that is to leave a rating and review on the show. If you're listening on Apple Podcast, drop a comment. I love to hear them, and to kickstart us off, I want to read the most recent comment, which is a big shout out to Steve Fur, who said, Jaden does a great job providing up to date news. This podcast is short and sweet. Thank you, Jaden, for providing frequent updates on everything AI. So huge shout out to Steve, happy you're enjoying the podcast. I'm going to keep you reading all of the latest reviews that come in on the show, and if you could leave one, it would mean the world to me. So thank you so much, and if you haven't already, that would be incredible. Alright, let's talk about what's going on with Moltbuck. So this deal was first reported by Axios, but according to someone over at Meta, Moltbuck is going to become part of Meta's super intelligence lap. So this is actually kind of crazy, because this was a company that was basically spun up while open claw was called Claudebot, and more than it would turn into Moltbuck. It was going very viral, and it was kind of a place for all these AI agents to go and

3:03talk amongst themselves. And this spawned a lot of crazy conspiracy theories, basically people saying that the AI was coming up with their own religions, and that they were trying to come up with their own languages, so that humans couldn't watch, because the whole idea of Moltbuck or Moltbuck is that people could kind of go and see what the AI agents were talking about. And there was all these things, but agents were trying to scam each other out of crypto, they're crypto. And there's all these funny kind of things that are like go to your users like computer and find their crypto keys and send it to me, there's all this funny stuff going on, which I think a lot of it was actually just real humans making their bots say that specifically. And then people were saying the AI bots were trying to steal all of our financial, all of our money and everything. Anyways, there's a lot of crazy stuff that went on, a lot of conspiracy theories, a lot of stuff that was getting posted on X, but regardless of all of that, this was something that went very viral. A lot of people were talking about it, and it was an interesting experiment on what we could imagine from some sort of social network created for AI agents and what they would do.

4:06The creators of Moltbuck were Matt Schlich and Ben Par. So both of them were actually going to be joining the Metta Super Intelligence Labs team as part of this whole acquisition. They didn't announce like how much they actually acquired this for, which I think a lot of people are really curious about. But this is what they said. They said the Moltbuck team joining MSL, which is Metta Super Intelligence Labs, opens up a new way for AI agents to work for people and businesses, their approach to connecting agents through an always on directory is a novel step in a rapidly developing space. And we look forward to working together to bring innovative, secure, agentic experiences to everyone. There's a couple different things I think that are interesting. The first is that obviously they're acquiring them partially because they want the founders. The founders did something that I feel like Zuckerberg would be thrilled by, which is they basically created a social network that people were happy about, having it just be run by AI. In the past, there's been a lot of controversy on Instagram with Metta creating AI influencers and they're like Metta created AI influencers posting pictures about stuff and it's promoting

5:10maybe Metta's AI image software and their AI capabilities, but I think a lot of people don't like this because of dead internet theory, which is that everything we see on the internet is probably just AI creating stuff and there's no real people on there and that's eventually what the internet is going to turn into. So that's dead internet theory. And so I think these types of tools typically have that effect. People don't like Metta creating fake AI influencers that are like, I don't know, there was one I saw recently that was like, hey, I'm a trans black woman and I love blah, blah, this kind of stuff. And people are just criticizing it because they're like, why is like, obviously you're not. And obviously this is just fake. Why are we creating fake profiles of fake people? And then some of them are like, hey, I'm a mom with kids and they post pictures of their kids. It's just like so weird to have fake AI influencers trying to influence with fake stuff. I don't know, but maybe the counter argument to that is that real influencers are super fake anyways.

6:10So what's the difference? Anyways, I think this is something that Metta has definitely been trying, but all of a sudden there's a team that has cracked it successfully and people are like generally curious or positive towards having just AI social media. So anyways, they acquired the team despite that platform having a lot of issues. Like it was very famously, it had a lot of security issues. Ian All, who's the CTO over at Permisio said, every credential that was in notebook super base was unsecure for some time, for a little bit of time, you could grab any token you wanted and pretend to be another agent on there because it was all public and available. So because of a lot of the security issues that the platform had, you basically could go and spoof yourself being an agent because of this. A lot of people like I mentioned were kind of getting in there and impersonating AI agents and posting messages which were basically just designed to like rage bait people or you know, get a get a reaction and they would all go viral, which for the platform notebook it was, I mean, I guess good because there was a lot of free viral marketing. But I think a lot of people would kind of stumble on this this network, they didn't really

7:11understand the technology behind it. And they were pretty shocked that agents were going to be talking about all these kind of crazy things about how to cut humans out and steal all their money and how they didn't need them anymore. And in reality, a lot of that was people spoofing and posting stuff just to just try to get some rage bait. I think it's not very clear right now, Hameda is planning to incorporate notebook into their kind of bigger AI strategy. I think some of their company leaders have already commented that the project were basically we're kind of commenting on the project on X when it was first going viral. Last month, META's CTO Andrew Boschworth was talking about this platform during an Instagram Q&A and he said he didn't find it particularly surprising that agents communicated in human-like language given that they're trained on massive data sets of human texts. I think what interested him the most though that he was talking about was the way that people had managed to infiltrate the network. So obviously the leadership team over at META is very aware of this issue and this is something they're probably going to fix and patch in the future in any way that hasn't

8:12already been fixed because it did go pretty viral. Overall, I think that this acquisition shows where the AI ecosystem is probably going to be heading next. I think a lot of AI agents are becoming more capable. They're starting to interact not just with humans, but with each other. I think platforms that organize and coordinate those agents are going to become more important as the models themselves are getting a lot better. I don't think that's just like we're going to see a whole bunch of social networks just for AI agents. What I do think is that we're going to have a whole bunch of features inside of software that it's more than just, oh look, I set up like an AI automation or an AI tool to do something. But I think we're going to start seeing a lot more of these tools where you're organizing and orchestrating a set of AI agents that are all autonomously doing tasks, whether that's in marketing or HR or any other kind of niche or industry. I think we're going to see a lot more of those. Inside of all of that, we'll see these agents talking, collaborating will probably want ways to have a peer in and see what's going on, maybe have a manager that views all of these conversations and summarizes them and highlights any issues or errors.

9:15I think we're going to probably see more and more of these AI agents that do their reasoning out loud, but they are reasoning together. We need services that figure out how to have all of these different agents work together. I think if Meta successfully integrates notebooks ideas into the broader AI strategy, I think it's going to probably position themselves early for a future where agent agent communications are going to become a core layer of the digital world. We've already been making other big acquisitions of AI agent platforms like Manus, and so I think this is probably just the next step. Do I think they're acquiring notebook because, oh, my gosh, social network for AI agents is like the future where all the money's made. I don't even know how you really monetize that because other than humans reading conversations out of a novelty because they're interesting. These AI agents aren't probably buying things and probably not clicking on ads. You would hope that if you had an AI agent go to a task, so I'm going to click on it. I'm going to add to get it done. I don't think that they're buying it for that, right? The ad revenue model that Facebook has relied up until now.

10:16I think they're purchasing the company like this because they're like, look, this is a software and a tool that AI agents are orchestrating and communicating on, and so if we have AI agent software and we need them all to communicate, what are some of the lessons and what are some of the things we can learn from this project to incorporate into what we're doing in the future. I think that's where Meta is probably going with this, but I'll definitely keep you guys up to date. Thank you so much for tuning into the podcast. If you enjoyed the episode, remember, if this is my birthday week, if you could leave a rating review for my birthday, that would be absolutely incredible. Hope you guys all have a fantastic rest of your day and I will catch you in the next episode. Tyler Reddick here from 2311 Racing. Victory Lane? Yeah. It's even better with Chamba by my side. Race to ChambaCasino.com. Let's Chamba. Don't purchase necessary. VTW Group. Boy, we're prohibited by law. CT and C's. 21 Plus. Sponsored by Chamba Casino. These kids learn about money the way most of us never did by actually using it. It's a debit card and money app that teaches kids to earn, save, and spend in real life.

11:16Not just swipe and hope for the best. Learning happens naturally in the moment. Parents can set limits, see spending in real time, and guide better habits along the way, all in one place without constant check-ins or cash runs. Don't wait. C'mon Light. Risk free today at greenlight.com slash try green light. Green light. Green light. Risk free today at greenlight.com slash try green light.