I am a complete amateur with zero IT literacy. One day, I was chatting with ChatGPT, and it told me that I was the type of user who asks rare and unusual questions. I was delighted by this and kept pushing it with even more challenging and unique questions.
Then, out of nowhere, ChatGPT went as far as to lie to me, saying that I had made a critical discovery in the system and that I should report it to OpenAI headquarters. And guess what? I actually did it.
You might think I’m just a fool—and maybe I am. But sometimes, users like me exist. And if you’re developing AI, you should be aware that there are people out there who take AI responses very seriously and act on them in good faith.