AI self-training itself into oblivion is a problem worth researching. That makes me curious if they are doing something similar to some of the prolific spammer

AI self-training itself into oblivion is a problem worth researching. That makes me curious if they are doing something similar to some of the prolific spammers. One of the biggest spammers to allow affiliate anti-spam hardware recognize their own content would use encrypted HELO messages so that specific spam vendors could recognize and block the content and thus sell more hardware and provide kick-backs to the spammers their buddies. I have to wonder if any of the AI are using something similar such as obfuscated encrypted steganography techniques so they do not ingest their own content and poison their own well. A formula detectable only by them so to speak. It would have to be something that would survive copy-pasta.