That "as a large language model" chatGPT kicked out for you at the end is boilerplate. There's a veneer of boilerplate responses on top of the underlying engine, and they patch that veneer to cover a new situation where the underlying engine does something problematic. But that doesn't fix the underlying engine. And it's not hard to work around the veneer.
That "as a large language model" chatGPT kicked out for you at the end is boilerplate. There's a veneer of boilerplate responses on top of the underlying engine, and they patch that veneer to cover a new situation where the underlying engine does something problematic. But that doesn't fix the underlying engine. And it's not hard to work around the veneer.