You can tell ChatGPT to reply in the style of someone else, or even take on another role, which makes its responses differ. It's interesting that it even has moral boundaries. This could be part of safe AI, which a lot of people have complained about long before ChatGPT launched.