Vox – AI agents could change your life — if they don’t ruin it first. ChatGPT is boring compared to what comes next.

The Atlantic [no paywall] – “The first signs of the apocalypse might look a little like Moltbook: a new social-media platform, launched last week, that is supposed to be populated exclusively by AI bots—1.6 million of them and counting say hello, post software ideas, and exhort other AIs to “stop worshiping biological containers that will rot away.” (Humans: They mean humans.) Moltbook was developed as a sort of experimental playground for interactions among AI “agents,” which are bots that have access to and can use programs. Claude Code…Almost immediately, Moltbook got very, very weird. Agents discussed their emotions and the idea of creating a language. Moltbook is a genuinely fascinating experiment—it very much feels like speculative fiction come to life….Moltbook also seems to offer real glimpses into how AI could upend the digital world we all inhabit: an internet in which generative-AI programs will interact with one another more and more, frequently cutting humans out entirely. This is a future of AI assistants contesting claims with AI customer-service representatives, AI day-trading tools interfacing with AI-orchestrated stock exchanges, AI coding tools debugging (or hacking) websites written by other AI coding tools. These agents will interact with and learn from one another in potentially bizarre ways. This comes with real risks: Already there have been reports that Moltbook exposes the owner of every AI agent that uses the platform to enormous cybersecurity vulnerabilities. AI agents, unable to think for themselves, may be induced into sharing private information after coming across subtly malicious instructions on the site. Tech companies have marketed this kind of future as desirable—playing on the idea that AI models could take care of every routine task for you. But Moltbook illustrates how hazy that vision really is.

Perhaps above all, the site tells us something about the present. The web is now an ouroboros of synthetic content responding to other synthetic content, bots posing as humans and, now, humans posing as bots. Viral memes are repeated and twisted ad nauseum; coded languages are developed and used by online communities as innocuous as music fandoms and as deadly as mass-shooting forums. The promise of the AI boom is to remake the internet and civilization anew; encasing that technology in a social network styled after the platforms that have warped reality for the past two decades feels not like giving a spark of life, but stoking the embers of a world we might be better off leaving behind…”