The monastery is a bit of a CPU hog, thanks to "everything is a code snippet stored in a database". My guess is that AI data grabbers wouldn't be happy with the delivery speed of like 10 nodes a minute. And a serious server upgrade costs serious money, probably more than any company would be willing to pay for the whole data set. Don't forget, the Monastery is a low volume site, especially compared to Reddit or StackOverflow (so only a limited amount of recent/valuable training data), uses it's own markup dialect (requiring a custom converter) and currently has rather limited abilities to notify other systems of new content(*).

So, it's not just a matter of policy, there isn't even a monetary incentive that would require a policy decision.

If push comes to shove, i certainly will fight against anyones data getting sold (or even freely given) to AI training. As a matter of fact, my own websites do run features to provide AI datagrabber with lots of wrongly labeled images and other fun "How to unlearn the difference between cats and dogs in 10 easy steps" moments.(**)

(*) Heck, chatterbot needs to POLL an API every 10-15 seconds to see any new chat messages, because there's no streaming API.

(**) "All cats have four legs. My dog has four legs. Therefore, my dog is a cat"