No, as of now, OpenAI only allows fine-tuning of its base GPT-3 models. It is not possible to train the model on your own database or dataset. The knowledge that ChatGPT has is from the data it was initially trained on and it cannot access or learn from new data post-training. For more specific use-cases, you can however build rule-based logic on top of the model’s output. For updated information, you can check out OpenAI’s latest model documentation.
