Hi, I'm researching LlamaIndex and want to use it in a company with in-house documentation hosted on Confluence. As I read through the documentation, I can use LlamaIndex without LLM but for some indexes LlamaIndex still uses LLm to create embeddings and summaries. Am I right or not? And is this a security risk if I have sensitive information? Thank you for clearing things out for me
It depends on the LLM provider. OpenAI, for example, as of June 14 2023 says they " will not use API data to train OpenAI models or improve OpenAI’s service offering." (ref https://openai.com/policies/api-data-usage-policies)
There's always risk - do you trust them to adhere to their policy? Is their network and infrastructure sufficiently secure to protect your data? Does your company have other legal obligations to protect user data that may require additional security guarantees from 3rd parties like OpenAI prior to transacting data with them?
Your company's legal department should have a policy that spells out what you can and can't do.
I applaud you for looking into this. I worked for a company that used Confluence and it was next to impossible to find anything. Everybody writing random pages with no structure or organization. Llama-index and an LLM would certainly help to find things easier 🙂
