I am a paying subscriber of ChatGPT-4o and have been using it alongside Wolfram for enhanced accuracy. While I understand that AI has limitations, I am disappointed by the lack of improvement in Wolfram’s performance, even after repeatedly correcting it with the right answers.
I would appreciate insights on this issue and any steps that can be taken to enhance the model’s responsiveness to corrections.
Thank you for your time.