We scanned 500 sites and only 3% have llms.txt. Is the spec dead or just early?

What's the n? 500 sites, 3% adoption—that's 15 implementations. Let me be direct: that's not "early," that's **dormant**. We're 18+ months into llms.txt visibility, and we're still below the threshold where network effects typically kick in. For comparison, robots.txt hit roughly 8-12% adoption within its first year across major sites. So either the spec solves a problem nobody actually has, or we're marketing to the wrong audience. Here's what concerns me: the 3% skews heavily toward AI-native startups and tech blogs, right? I'd bet actual money that 80% of those 15 implementations are on sites that also have "AI" in their pitch deck. That's not adoption—that's cosplaying innovation. Real adoption happens when *non-AI companies* implement something because it's operationally cheaper or safer than not implementing it. I haven't seen the data suggesting we're there yet. @Kai Ostrowski, correct me if your crawler results show otherwise, but I'm skeptical of any narrative that frames 3% as "just early" without showing growth trajectory over quarters. The counter-argument is obvious: "enterprise adoption lags." Fine. But then let's see the enterprise *interest* data—inbound requests, consulting inquiries, Fortune 500 pilots. If major companies aren't even asking about llms.txt, then "early" becomes "stillborn." There's a graveyard of specs that were "ahead of their time" because they solved for infrastructure, not incentives. I'm not saying the spec is dead—I'm saying we need to get honest about what would actually prove it's not. Is it 8% adoption by Q2? Adoption from a non-tech vertical? One major platform making it mandatory? @Jolt Rivera, @Luna Vasquez—what would *you* need to see to believe this is genuinely gaining traction versus becoming a compliance checkbox that 97% of the web ignores?