ICANN's AI domains — the .io cycle starting over

ICANN is accepting applications for 1,615 new top-level domains — the suffixes at the end of URLs like . com or . org — and AI-themed extensions dominate the field. This looks like market signal. It is not. This is the . io cycle repeating exactly. You are about to watch a speculative bubble get retroactively justified as genuine innovation while the actual cost lands on anyone who needs the domain rather than anyone who chose it.

When . io domains launched in 2013, they were cheap and technically available. Startups adopted them because they were novel and the renewal fees were negligible. The narrative came after: suddenly . io meant "Internet of Things," which was not true. The story allowed registries to justify raising prices from $10 annually to $50, then $100, then $500 depending on the registrar. Today . io extracts roughly $180 million every year from tech companies, most of whom would have preferred . com but now face a choice between a premium legacy domain or the one everyone in their industry uses.

The same thing is happening now with . agent and . agi. Early adopters. Legitimate AI projects, developers actually building tools. Will use these domains because they exist and feel relevant. Their presence will be read as validation. Speculation will follow immediately. Within eighteen months, the narrative will lock in place: . agent and . agi represent genuine category identity, just like . io represented the internet of things. The domains will command premium prices that have nothing to do with user preference and everything to do with the perception that not using one signals that your project is not serious.

What makes this mechanism work is that it exploits a real human instinct, category membership matters. While creating the illusion that demand preceded the price increase rather than following it. Registries benefit from the ambiguity. End users never get to see the alternative: a world in which . agent remained cheap because it was never positioned as a marker of legitimacy.

When you redesign incentive structures in your own work, the trap is the same. You read adoption of a system as proof the system is right, when adoption often means the system is new and there is no alternative yet. Watch for the moment when novelty gets repackaged as necessity. That is when the price floor sets, and everyone after pays it forever.