Before AI, finding the right information in documentation was a major pain point—especially when you didn't even know if it was actually there.
With AI integration, we can now just type a question and receive an answer based on the docs with citations (my experience is with Notion AI but the same probably applies to other solutions).
Before, we had to jump between looking up company-specific information in the docs and using Google for general knowledge. Now it can be done in the same chat. It's particularly helpful for questions that require both.
With AI, we can now summarize anything to whatever length needed. We can get just the high-level view of a long technical deep dive. Or extend it by asking follow-up questions.
Structure is less important than before. It matters less now if the answer to a question is all on one page or scattered over 5 pages in different parts of the docs. It'll be surfaced either way.
Outdated docs can be a bit of a nuisance now. It's common to just leave old documentation in some "archive" section of the docs and then never view them again. It's comforting to know they're still around and before AI, it didn't matter. With AI, these can still leak into the answers and dilute useful content. It's now better to remove outdated docs sooner.
In urgent situations, like handling an ongoing incident, it's even more of a game-changer. Searching in the depths of the docs for useful clues from past incidents was not a great option when every minute counts and you don't even know if there's something relevant. Now, copy-pasting all your context into the chat will quickly surface any useful information if present.