I've seen many of these "AI" offerings regurgitate their training data verbatim. If chatgpt were miraculously immune to this, despite taking the same basic app

I've seen many of these "AI" offerings regurgitate their training data verbatim. If chatgpt were miraculously immune to this, despite taking the same basic approach as everybody else, and without significant efforts to prevent regurgitation, then the onus is on them to prove that.