What happened
Ahrefs analyzed roughly 137,000 domains and found that about 97% of published llms.txt files received zero requests. The llms.txt convention was proposed as a way for sites to guide AI systems to relevant content, but the data suggests almost none of the files are actually being fetched.
Among the files that were accessed, most requests came from SEO tools, scanners and validators rather than AI retrieval bots. Actual AI systems accounted for only a small share of the traffic, undercutting the idea that publishing the file drives AI citations.
Why it matters
The findings indicate that adoption of llms.txt is outpacing any demonstrated utility. Teams spending effort maintaining these files may be optimizing for a channel that AI systems largely ignore.
It is a useful reminder that emerging AI-optimization conventions should be validated with data before they become standard practice.
MintedBrain take
Before treating llms.txt as a meaningful AI-optimization tactic, teams should confirm whether AI systems actually request it for their sites. The convention may still gain traction, but the current evidence argues for caution rather than broad investment.
Discussion
Sign in to comment. Your account must be at least 1 day old.