Is ChatGPT Getting Worse?
A new study (PDF) from Stanford found that ChatGPT performed worse on certain tasks in June than its March version. The paper supports a widely held, though unproven, notion that the AI language model's performance in coding and compositional tasks has deteriorated in recent months. Fortune reports: The study compared the performance of the chatbot, created by OpenAI, over several months at four "diverse" tasks: solving math problems, answering sensitive questions, generating software code, and
Read more »