What we look at
- The code
- Traces left by AI site builders and AI coding tools such as Lovable, Bolt, v0 and Durable, plus WordPress plugins that write posts automatically.
- The writing
- Words chatbots overuse, em-dash habits, "it's not X, it's Y" phrasing and sentences that all run the same length.
- The images
- File names and hosts that give away DALL·E, Midjourney, ChatGPT and other image generators.
- The publishing
- Hundreds of pages appearing on one day, or a huge site on a domain registered a few months ago.
Questions
How can you tell if a website is AI-generated?
No single clue proves it. The strongest evidence is code left by AI website builders. Writing style (AI-typical words, very uniform sentences), AI image file names and mass-publishing patterns add up when several appear together. This tool checks all of them at once.
Is this checker accurate?
It returns a probability, not a verdict of guilt. Builder fingerprints are strong evidence. Style signals are weaker alone because people also write "seamless" and "robust". Treat a high score as "worth a closer look".
Why does the confidence say "Low"?
Either the site had very little readable text, or it renders its content with JavaScript, or it blocked part of our requests. We only score what we could actually see.
Does Google penalize AI-generated websites?
Google says it rewards helpful content regardless of how it is produced, but mass-produced pages made mainly to rank are treated as spam. That is why buyers of websites and SEO agencies check sites before purchase.
Do you store the sites I check?
No accounts, no stored history. Pages are cached for up to one hour to make repeat checks faster.