Hello, OpenAI Support Team!
I am using a customized bot based on ChatGPT, but I have encountered a significant issue: each time I update the instructions or add new files to the knowledge base, the bot loses memory of previous errors and corrections.
This is critical because the bot should learn and adapt, but in practice, it resets every time, even if I have already corrected its mistakes. This memory reset contradicts the logic of an AI that is supposed to improve progressively.
How it should work:
The bot should remember past mistakes and corrections, even after updating the knowledge base.
If something needs to be forgotten, it should be specified manually, rather than happening automatically.
Adding new data and adjusting instructions should not erase knowledge of previously made corrections.
I kindly request that you fix this issue or add an option to retain the correction history when updating instructions. This would greatly enhance the user experience and improve the bot’s efficiency.
Thank you! I look forward to any updates or explanations regarding this matter.
Best regards,
[Dimas]