Chatgpt on unstructured data

I want to train/finetune ChatGPT on my set of unstructured data namely pdf’s, ppt’s, doc files.

The aim is to get Chatgpt responses from above data in context.

Does OpenAI provide provision to train Chatgpt via API ?

The ChatGPT AI cannot be fine-tuned.

You’ll want to use something like GitHub - jerryjliu/gpt_index: LlamaIndex (GPT Index) is a project that provides a central interface to connect your LLM's with external data.

Basically, you will use embeddings to manage your unstructured data and provide it in the ChatML you submit to the ChatGPT API.