The worst part about this is that if there is another set of bots that tries to generate engagement, then the training data isn't coming from humans either. You have one set of actors spamming. And another set of actors upvoting stuff, predominantly their own but maybe also other random posts. So the resulting posts don't necessarily even cater to humans. It will be real online hellscape.
The web will transition strongly to verified identities, like we have with SSL certs. Along with filtering out people who use AI to post under a verified identity and get caught, It’s the only way to help ensure you’re reading actual human content.
That is how legally admissible e-signature schemes work. When the certificate holder dies or the certificate expires, it cannot any more be used to sign further documents.
