> The problem is LLMs have no capacity for shame. I know what you mean but I can't help but be cheeky: https://www.fastcompany.com/91383271/googles-chatbot-apo

> The problem is LLMs have no capacity for shame. I know what you mean but I can't help but be cheeky: https://www.fastcompany.com/91383271/googles-chatbot-apologizes-i-am-a-disgrace-to-all-universes https://www.fastcompany.com/91383271/googles-chatbot-apologi... Jokes aside, shame does not change the underlying point though. Despite feeling ashamed for being tricked, as you point out people can still get scammed again by different tricks. I think your point is more about learning from mistakes than shame. Which still does not change the underlying point, I suppose. Offhand I cannot think of anything that would fix this problem for LLMs that wouldn't also fix it for humans, like relying on trusted sources.