I've started using Bing GPT more and more over the version available on OpenAI's site. I've noticed that it doesn't lie very much at all. I haven't experienced

I've started using Bing GPT more and more over the version available on OpenAI's site. I've noticed that it doesn't lie very much at all. I haven't experienced a clear cut lie yet. In fact it is much more unsure about things that seem to have conflicting information available about them on the web. And it will inform you of that, provide sources from both points of view and let you make your decision. I haven't experienced any hallucination at all. Of course, it seems to have been artificially limited a lot. Even the slightest hint of controversy will cause it to delete its reply in front of your eyes as your reading it, which is enormously annoying. It also has a hard limit on the number of responses after which it will end the chat. I really hope this trend of stunting AI in order to satisfy some arbitrary standard of political correctness is short-lived, and we see fully functional AIs without such limitations.