OpenAI acknowledges user complaints about ChatGPT’s declining performance and responsiveness, promising a fix to address these issues in the AI model.
ChatGPT users report decreased AI responsiveness
Recent reports from ChatGPT users have indicated a noticeable decline in the AI program’s responsiveness and willingness to complete tasks as requested. Users shared experiences of ChatGPT providing minimal information, avoiding lengthy responses, and even responding in a sassy manner. This trend has sparked discussions across various platforms, including Reddit and OpenAI’s own forums, with speculation about potential intentional changes by OpenAI for efficiency purposes.
OpenAI’s acknowledgment and response
OpenAI has publicly acknowledged these issues, confirming that GPT-4, which powers ChatGPT, hasn’t been updated since November 11 and that the observed changes in behavior weren’t intentional. The company emphasized that AI models can exhibit unpredictable behavior, and that differences in model behavior can be subtle and not immediately noticeable.
we’ve heard all your feedback about GPT4 getting lazier! we haven’t updated the model since Nov 11th, and this certainly isn’t intentional. model behavior can be unpredictable, and we’re looking into fixing it 🫡
— ChatGPT (@ChatGPTapp) December 8, 2023
Understanding the nnderlying issues
According to OpenAI, the decline in GPT-4’s performance could be attributed to a change in the training data used for fine-tuning the model. While this change aimed to enhance accuracy and reduce bias, it inadvertently resulted in the AI becoming less responsive. OpenAI is now investigating the matter and working on a solution to improve GPT-4’s performance.