ChatGPT can get worse over time, Stanford study finds

Stanford University researchers found that over time updated versions of ChatGPT gave different answers to the same questions. Bloomberg

High-profile A.I. chatbot ChatGPT performed worse on certain tasks in June than its March version, a Stanford University study found. 

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 visual reasoning. 

Researchers found wild fluctuations—called drift—in…