Copilot for Microsoft 365 Tech Accelerator
Feb 28 2024 07:00 AM - Feb 29 2024 11:00 AM (PST)
Microsoft Tech Community

Do Copilot reference documents need to be stored on OneDrive?

Copper Contributor

I understand that Copilot retrieves information from my documents to respond to prompts (Retrieval Augmented Generation - RAG).  

Do my documents need to be stored on OneDrive in order to be indexed and retrieved?  

What if I have relevant documents stored on my hard drive or a local network server - will Copilot index them and retrieve information from these documents as well?  (if so, how?)

6 Replies
Nope, the data needs to be stored in Microsoft 365 storage (so onedrive/sharepoint/teams/outlook) to be indexed by semantic index and retrieved using RAG
Not totally agree :
Local files: Connect to files in your Azure AI portal, providing flexibility and convenience in connecting your data. We ingest and chunk the data into your Azure Cognitive Search index. File formats such as txt, md, html, Word files, PowerPoint, and PDF can be utilized for analysis and conversation.
https://startups.microsoft.com/blog/8-steps-to-building-an-azure-openai-copilot-for-your-startup/
This is not something to do with Microsoft 365 Copilot, this is a separate stack on Azure.
yeap but with graph connect we can plugin into Semantic Index
Last I checked, Cognitive Search and Azure OpenAI is not supported as a connector yet with Graph Connect or is there a new integration possible? https://www.microsoft.com/microsoft-search/connectors/?publisher=Microsoft&category= or do you mean building a custom connector?

@LUDOVIK DOPIERALA consumers going to https://copilot.microsoft.com/ don't have the option to customize data sources though, right? As a user paying for Copilot Pro and going to https://copilot.microsoft.com/ the "attach file" icon shows only files in OneDrive. I'd be glad to be wrong, because that seems a pretty big limitation.