Can AI assistants read my website?
Assistants use separate crawlers for training and for search. OpenAI, for example, uses GPTBot to collect training data and OAI-SearchBot to find pages for ChatGPT search; Perplexity uses PerplexityBot for its answers. Your robots.txt file can allow or block each one. Some site builders and security tools block them by default; blocking a search crawler can leave you out of that assistant's answers without your knowing.
Blocking them is a legitimate choice if you do not want your content used. It should be a decision, not an accident.
How do I tell machines who I am?
With structured data: a small block of code in the page that states facts in a standard vocabulary (schema.org). Say what the organisation is, and describe the page with the type that fits, such as Product, Service, Article or FAQPage. This removes guesswork about what your business is and does.
What kind of content gets quoted?
Assistants lift short, direct answers. Make them easy to find.
- Use real customer questions as headings.
- Answer each one in the first two or three sentences beneath it.
- State facts plainly: prices, sizes, times, places.
- Keep the text in the page itself, not built by scripts after loading.
Do I need an llms.txt file?
It is optional. llms.txt is a proposed plain-text summary of a site for AI assistants. It costs little to add, but there is no solid evidence yet that it changes how often a site is cited, and Google has said it does not use it. Treat it as a small extra, after the steps above.
Can anyone guarantee I will appear in AI answers?
No. Nobody outside those companies can see or control which sites an assistant quotes. These steps remove the obstacles that are known to keep a site out.