Is anyone surprised? By definition LLM can’t be 100% correct and LLM hallucination poses significant challenges in generating accurate and reliable responses. ChatGPT Answers Programming Questions Incorrectly 52% of the Time: Study. To make matters worse, programmers in the study would often overlook the misinformation. https://gizmodo.com/chatgpt-answers-wrong-programming-openai-52-study-1851499417
@nixCraft It's sad to see so many people trying to justify the "toy" by blaming the people who asked the question for the wrong answer provided by the AI.