Catherine Bell argues that ChatGPT can be taught and coached through prompts to change its outputs, unlike Google which just knows things. She shows how conversation and prompt engineering let you shape what the AI produces.
Yeah, I'm not. I'm not saying that that chat GPT works 100 every time. You have to play with it a little bit, coax it a little bit, coach it a little bit.
Yeah. Understand what you're.
Doing. It's called learning, not knowing. That's what makes it different. Google knows, but you can't teach Google. Whereas ChatGPT, you can teach it and it'll learn and then know forever.
Well, and to the point, like, I can give it a prompt to create an image. You give it a totally different prompt to create an image in yours. Like from mine to yours. Right. Completely different out outcomes come from it. Because of all the conversations I've been having and the different conversations you've been having.
