Access to company's proprietary data for ChatGPT Learning

[timminer]

Need an architecture for safe implementation of ChatGPT into the enterprise architecture of a financial services firm. The firm wants to offer a single app interface that users can query using a LLM, but the data used to create an answer is sourced both from OpenAi’s massive data set as well as the firm’s internal, proprietary data. The difference between these two data sources should be transparent to the user and a single LLM answer should be returned.
How can ChatGPT’s model learn and train from the internal company data without this data being exposed to the rest of the world?

[sps]

Welcome @timminer

Training private version of model

All models that are fine-tuned are private to the account that requested the fine-tune. Only the account that requested fine-tuned model has access to it and controls access to it, and yes you get a whole new fine_tuned_model like ada:ft-your-org:custom-model-name-2022-02-15-04-21-04 I would definitely recommend reading through docs on fine-tuning and embeddings before you proceed. On a first look, it appears like embeddings could do the job for this case, but depends on the details. Do yo…

[nnarula]

Hi, I am looking to solve the same problem. Were you able to resolve it and create a solution?