Mike Mann recently shared some thoughts with me, claiming .AI is "presumably at its peak." And you know what? Only Mann can say something like that and get half the industry nodding along while the other half loses their minds.
His logic is pretty straightforward: there aren't enough profitable AI companies to justify what everyone's paying for these domains. It's speculation on top of speculation, and when the music stops, a lot of people are going to be holding worthless names.
I get it. But I'm not sure he's right.
Look at the actual sales data. Prediction.ai went for $109K. Agent.ai sold for $125K. Automate.ai hit $100K. These aren't domainers selling to other domainers and patting each other on the back. And it's not just sales. Real companies are building on .AI. Perplexity.ai is worth over a billion dollars now (the company, not the domain). Character.ai, Synthesia.ai, and dozens of venture-backed startups picked .AI as their primary domain. That's not nothing.
So where's Mann coming from? He's right that most AI startups today are going to fail. The whole sector is inflated with venture money, and half these founders are doing "AI-ish" things without any real business model. That's creating artificial demand for .AI domains, and when the funding dries up, which it always does, a lot of those renewals are going to hurt.
The lower end of the .AI market is absolutely a bubble. People are hand-regging garbage thinking they've found gold. They haven't. If your portfolio is full of long-tail .AI names that sound clever in a text message but would never work on a business card, Mann's talking to you.
But here's where I think he's missing something. .AI isn't like .mobi or .club or whatever hyped extension failed last time. AI is a platform shift. It's one of the few real ones we'll see in our lifetime. And .AI isn't some clever branding trick—it literally names the industry. There's no explanation needed. It just works.
Even if thousands of AI startups die, the winners are going to be massive. And those winners are the ones buying premium .AI names right now. Perplexity and Character aren't building billion-dollar companies on .AI by accident. They chose it because it fits. If those companies succeed and some of them will, the extension isn't peaking. It's maturing.
Premium .AI names don't need the entire AI startup ecosystem to survive. They just need the handful of giants that are going to define the next decade.
So yeah, there's a bubble. But it's not the bubble Mann thinks it is. Ultra-premium one-word .AI domains? Those aren't going anywhere. Solid two-word combinations? Case by case, but there's real value there. The hand-reg noise everyone's sitting on? That's the bubble. Most of it will drop.
Mann's right that speculation is overheated. But I don't think .AI is at its peak. The hype cycle will cool off, sure. The garbage will flush out. But the extension itself has a long runway ahead.
If you're buying .AI, buy what real companies actually want. Not what sounds futuristic in a Discord channel. The winners will stick around. The rest won't.