ChatGPT, the AI chatbot, may not be the best resource for software engineering queries, according to a Purdue University study. The research found that 52% of the chatbot’s answers to Stack Overflow questions were incorrect, and 77% were verbose. Despite the comprehensive nature of the responses, users often overlooked incorrect information due to the well-articulated answers. The study highlights the risk of misinformation spread by chatbots and suggests reconsidering the use of ChatGPT for such prompts.
Read more at ZDNET…