I don't think you need AI but you do need to think about the audience for the printed catalogue. A few years ago I did the website for a retailer of clothes fo

I don't think you need AI but you do need to think about the audience for the printed catalogue. A few years ago I did the website for a retailer of clothes for the elderly, and they were doing it old school with the catalogue, printed order form and excellent customer service by phone. Their niche was the demographic every other clothing retailer avoided. Unless you have a similar niche, you have to ask about whether a printed catalogue is worthwhile. AI could potentially help but how do you plan and budget time for that? It could take anything between two minutes and two years to get right. Meanwhile you could do it old-school with artworkers slaving away. Alternatively, you could automate the process to use print stylesheets where you specify the page size and then populate the content with CSS grid layout. The printed catalogue could then be created on demand (and cached) so that it automatically updates itself. This could be a manageable process that you could plan and budget for. In your product database you could have fields for layout preferences so that you can specify the featured product for each page and what to downgrade in the presentation. I would say this is definitely viable and one reason this is not done is that any company still invested in print catalogues will have an artworker department and nobody in such a department would invest time into automating their job.