The edit we run on every AI draft

What Google actually says (checked 4 November 2026)
Google doesn't ban AI-written content. Its guidance is that content should be helpful, reliable and made for people, however it's produced. What breaks its rules is generating lots of pages mainly to manipulate search rankings, with AI or any other way ("scaled content abuse"). It also says to check AI output for accuracy before publishing, including titles, descriptions and alt text.
- Using generative AI content: https://developers.google.com/search/docs/fundamentals/using-gen-ai-content
- Creating helpful, reliable, people-first content: https://developers.google.com/search/docs/fundamentals/creating-helpful-content
- Spam policies, scaled content abuse: https://developers.google.com/search/docs/essentials/spam-policies#scaled-content
- Google's 2023 post on AI-generated content: https://developers.google.com/search/blog/2023/02/google-search-and-ai-content

So the question isn't who typed it. It's whether the page says anything a reader couldn't get from every other page on the topic. That part is usually what only you know.

THE CHECKLIST: before any AI draft goes live
1. A real price, or a real price range, and what makes it go up or down.
2. A real job you did: roughly where, what was wrong, what you did, how long it took. Ask the customer first if they could be recognised.
3. A photo you took: the job, the team, the premises. Not stock.
4. A customer's actual question, in their words, and your answer.