Hello, Could it be that the AI is getting dumber and dumber as we are closing in into the release of GPT-5? I feel like the reponses in the past few days are geting less context knowledge and more randomness
#Dumber
autumn grail
fickle wingBOT
Hi @autumn grail. I'm kapa.ai, your friendly support bot. I will try to answer your question by searching through the knowledge base. :wave:
Based on the knowledge sources provided, there are indeed some users who have reported a perceived decline in the performance of their GPT models. For instance, one user reported that their models started producing responses that were disjointed and irrelevant to the context of the conversation, and the responses had become repetitive source (https://community.openai.com/t/599319). Another user mentioned that the coding ability of the models seemed to have worsened, with the AI hallucinating and getting rid of code randomly in its replies source (https://community.openai.com/t/691846).
However, it's important to note that these are individual user experiences and may not reflect the overall performance of the models. The performance of AI models can be influenced by a variety of factors, including the specific tasks they are being used for, the quality and relevance of the input data, and the way the models are instructed.
If you're experiencing issues with your models, it might be helpful to review the best practices for instructing models. For example, the advice for instructing gpt-3.5-turbo-0301 may not apply to future models source (https://cookbook.openai.com/examples/how_to_format_inputs_to_chatgpt_models#3-tips-for-instructing-gpt-35-turbo-0301).