Though interestingly, the observed difference in assessment suggests (though does not prove) that sampled AITA posters are not one of these models. I guess it’s

Though interestingly, the observed difference in assessment suggests (though does not prove) that sampled AITA posters are not one of these models. I guess it’s possible they have a very different prompt though…