Leadership is asking about AI. Someone at a recent board meeting said the organization should be doing something with it.
I would ask them exactly what needs to be done, not just what they want to be done. I would find out what their needs are, where the bottlenecks are, where they think they need it first, and evaluate whether that's really where they need it. They need to take a long hard look at what's taking too much time, and they also need to take a look at what could be automated.
Why bolting AI on doesn't work
Adding AI to a website is much more difficult if the website is not structured properly. Bolting on AI as an afterthought doesn't work as well and won't get the results that you want.
These are the things that usually get in the way.
Content that isn't structured consistently. AI can only summarize, tag or recommend what it can read.
No API layer. If the site has no clean way for other tools to connect, there's nothing for the AI to plug into.
Old versions and thin hosting. A site on end-of-life software can't take on the extra load.
Messy data. Duplicate and outdated pages produce unreliable answers.
The first thing I'd check
To check if a website is ready for AI, the first thing I would look for is content moderation, to make sure that they can handle that. You don't want to have AI content posted up without a human read on it. It's supposed to be an aid to help your editors, and you need to have phases that the content goes through, from unpublished to draft to review to published.