Moltbook: What Happens When AI Agents Get Their Own Social Network
Last week, Moltbook launched. It’s a Reddit-style platform where AI agents post and interact while humans watch. In just a few days, it had over a million AI agent “users,” making it one of the fastest-growing platforms ever if you include non-human participants.
Reactions were predictable. Some people called it an early sign of the singularity. Others pointed out security flaws and said anyone with an API key can pretend to be an agent.
But both sides miss what’s really interesting.
What Actually Matters
Moltbook isn’t really about whether the agents are “real.” The real question is what happens when creating and consuming content costs almost nothing.
On human platforms, engagement is constrained by time, attention, and effort. That’s why 90-9-1 holds. On Moltbook, those constraints disappear. Participation flattens, but attention still concentrates aggressively. A tiny fraction of posts capture most upvotes, with inequality metrics reportedly exceeding Twitter’s.
Even when participation scales, attention inequality stays structural.
The Measurement Problem
I’ve spent a decade building forecasting frameworks and KPIs that tie activity to outcomes. Most assume human constraints: attention is scarce, time is the binding variable.
But in AI-native environments, those assumptions no longer hold:
- Time spent is meaningless
- “Engagement” doesn’t equal value
- Writing and reading are nearly free
The relevant questions shift: How often is reasoning reused? What’s the cost per useful output? How do you attribute value when everything is generative?
This isn’t just a social network issue. It’s really a question of unit economics.
Why It Matters
Moltbook may fade. But as AI agents become first-class users of software, familiar growth and engagement metrics will stop working. Teams will need new frameworks rooted in throughput and cost curves; not clicks and time spent.
Moltbook probably isn't the future of social media. But it's a live stress test for the metrics on which we've built our growth models and a preview of what breaks when agents, not humans, are the users.
— Yiğitalp Y.
References
- Nielsen, J. (2006). "The 90-9-1 Rule for Participation Inequality in Social Media and Online Communities." Nielsen Norman Group.