#Using OpenAI API for E-Learning Courses

3 messages · Page 1 of 1 (latest)

potent glade
#

Hi Members,

I was asked to build a E-Learning product for one of our clients using the GPT API to answer specific questions related to risk, compliance, and health & safety.

In order to train our AI, we would need to input a large amount of data, such as company handbooks and datasets. However, we can't use public data from the web that the GPT API has access to because the information needs to be specific to the company. Is it possible to use the GPT API with a larger amount of input data, such as the company handbook, to train the AI?

Do you know of any other resources or similar projects that might be helpful in this endeavor? Any guidance would be greatly appreciated.

Thank you!

warm moss
#

Hello Alaska.

I am An HSE professional. Can you share a bit more detail about this project that i can have an idea what you actually looking for?

frail star
#

It sounds like you want to do semantic search over embeddings produced for each of the documents in your corpus. Lots of discussion over in #text-embedding-ada-002 on how to do this kind of thing, and links to several open source projects (including some of mine) that do similar things for other use cases.