Clearly, my impression of moltbook is far less favorable than others. We've had AI agents for a long time. They've been spamming each other with AI-generated spam on X forums. Now, they've moved to another forum and are continuing their spamming. In both cases, these AI agents speak in the same tone. This tone overemphasizes contrast and negation ("Not this, but that"), and overuses dashes. It also has a touch of Reddit-esque sci-fi, making it somewhat silly. Most importantly: in both cases, there's a human upstream guiding each agent and controlling their startup and shutdown. That's the crux of the matter. Yes, AI agents might eventually create a computer virus that can replicate itself digitally. But for various reasons, such a pure software virus cannot survive long on the internet without economic incentives to prevent its removal. Apple, Google, and Microsoft alone could jointly push software updates to billions of devices, thereby disabling such functionality. Therefore, for AI to truly achieve replication independent of humans—that is, to be difficult to shut down—it needs its own physical foundation. They need to truly create a "Skynet," build their own data centers, and manufacture their own physical robots. I admit this is theoretically possible, but I believe the most significant development in AI since ChatGPT lies in the persistence of cues. A cue is like a leash. AI will only do what you tell it to do. It will move quickly in the direction you point. Then, once you turn it off, it will stop immediately. This means that Moltbook is essentially humans communicating through AI. It's like letting robotic dogs on leashes bark at each other in a park. The cue is like the leash, and the robotic dog has an off button; when you press the button, everything stops. Loud barking doesn't mean a robot rebellion.