AI-powered
podcast player
Listen to all your favourite podcasts with AI-powered features
The Reverse Turing Test and Large Language Models
I was reading an economist art on large language models. The author said, well, I have to give credit to GPT three that helped me write this article. And it came along with two interviews, one by Blaz Diarka, and the other one by Douglas Hofstetter. And they were diametrically opposed. It has a very sophisticated understanding of theory of mind and social interactions. But if someone's interviewing it with nonsense, things that don't make any sense, well, what's it going to do? You ask a nonsense question to get a nonsense answer that summarizes, don't have a stutters interview. That is to say, instead of the human trying to detect