A new study has examined issues of large language models (LLMs) failing to flag articles which have been retracted or are discredited when asked to evaluate their quality.
New research suggests ChatGPT ignores article retractions and errors when used to inform literature reviews
