This is interesting on the subject. Maybe gives you some understanding. Posted on HN a few minutes ago.
Search engine crawlers are more mature and better written.
I suspect a lot of LLM crawling an development is done under time pressure to get things done while the investors money is still coming in to fund it. DO stuff in a hurry, and it will be less competently done.
Whenever the market decides the Internet is too full of slop to be usable for "training" the one that has the most copies of the pre-"AI" Internet wins. Some of the traffic is likely "AI" "tool use", i.e. bot scraping as part of running some LLM, i.e. "AI" "research".
The big scraping bots have gone from stupid to ruthless. Previously it was irritating that some of them got stuck traversing cyclical link paths on your site or on-the-fly generated pages, now it's like your silly family blog suddenly got very popular for no good reason and it puts a lot of load on the tiny amount of hardware it's served from.