I just hooked up my entire Google Drive to Chat GBT. And now I can speak to pretty much any of my files instantly. And unlike other tutorials, I didn't have to create a custom GPT with special custom actions. I didn't have to create a custom knowledge base or use chat projects. All I had to do was click a button, create a catalog of my entire Google Drive, and I can now speak to any of my files with no size restrictions. This is actually a feature that came out a few weeks ago in chatb Teams that not enough people are actually talking about. You can now turn your documents, Google Sheets, and even Google Slides into material that you can speak to and get accurate answers from. All you have to do is just tap internal knowledge, ask your question, and voila, you get four to five responses of possible matches with files living in your Google Drive. In this video, I'll walk you through exactly how to set up internal knowledge in minutes, as well as who this feature might be perfect for from both an operational standpoint as well as a security standpoint. Let's dive right in. All right. So, when you log into chat teams and again it's teams or enterprise. This won't work at least for now on the normal personal account. You'll be able to see a brand new toggle when you go on GB40 specifically called internal knowledge right here. When you click on it, it won't show much. It'll ask you whether or not you want to connect. And what it will ask you is whether or not you want to use OOTH, which is just authenticating your Google Drive to be able to create a catalog, which is like a big mind map that will continually sync with your latest files after you've done the initial sync. Assuming you don't have any sensitive files, then you can most likely just sync everything you want in your Google Drive. But if you want to be able to curate exactly what gets synced and doesn't get synced, I'll show you how you can do that as well. Once you actually initiate the connection, it'll take, depending on the size of your Google Drive, anywhere from 5 minutes all the way to 20 minutes, depending on the thickness and the different layers of folders and sub files that you have. In my case, since I only store my personal files here, including my YouTube stuff, as well as any experiments I do for the channel, it took around 10 to 15 minutes to finish the entire sync. Once it's all done, I created some fictitious files here called company files. And we have a few different folders. one that says content creation, one that says core files, and each one can be a pretty hefty file as well. They're not very short. This one's around 15 pages. And I've created some other fictitious documents, all using Manis AI just to help me out. So, this is some fake employee handbook for a madeup company called Nexitech Solutions. And if we go down here, this one's 11 pages. And we can keep going through the folder. It's pretty thick. And you can see here if we go to market research report, this one's around 24 pages and has all kinds of nested information. Now, let's say I want to ask a specific question from this document. Let's say about the market overview and size and it's about this market research report and I want to ask about the global enterprise AI software market size growth from 2020 to 2024. So, let's ask this specifically. Let's put that here. All right. And I'll say I want to be able to search any market research reports that you can find that'll have information about the following. And then I'll just paste that exactly. And then it should now reference and read documents basically doing a mini Google search within my Google Drive to find the closest matches to my query. And just like that, we get the identical response from that Google doc right here. So we can see 2020 says 6.8 8 billion, 8.5 in 2021, 10.2 in 2022, 12.7, and then 15.8. And if we go back to the Google doc we referenced, we see we have the exact same numbers there. So 10.2, 12.7, and 15.8. And it also goes to the next part of the document where it goes through the market forecast from 2025 to 2030. And at the very bottom, usually let's say you're using deep research, it would show you the websites it searched or typically if you use the knowledge base, it would say I reference this PDF. In this case, when you click on sources, it's only way it can reference knowledge is from the Google drive. So it tells you exactly which documents or in this case document it really got this answer from. And then it also referred to marketing strategy and enterprise AS software agreement to see whether or not there would be a match with our question. So you can see just like that you now have a powerful mini rag system at your fingertips without having to go through all the hoops of connecting custom GBTs to special tools or doing something like chatt projects. Now how do you know whether or not this feature is perfect for you? So if we go into their mirrorboard right here now in chat teams you can obviously use this feature pretty easily and it's typically optimized for teams of 10 or less. And the reason why is when you get into access issues as to whether or not everyone can have access to the same knowledge base. This is where it becomes more complicated at scale. Now there is an entire section I'll show you where you can manage what files different people have access to. But again if you have a company of 50 100 people that's where it might make more sense to look at chatd enterprise. But before we get ahead of ourselves, in terms of the types of files you can sync, you can do Google Docs, Slides, PDFs, Word documents, PowerPoint, and plain text files. So this whole mix is eligible. Where it struggles from my experience is if there are images within PDFs, it's still not as good at say as clawed when with its PDF analysis. So that's one thing to keep on the lookout for. Now, if we want to be able to manage usage, we can go to here and then go on to manage workspace and then go into connectors. And then if we go here, you'll see that you have the connection type, you have the domain name, and then you have something called user access where you can decide who in your workspace has access to the specific components of the Google Drive. And if you want to block syncing whatsoever from specific items or URLs of documents, you can see this section here says you can block items here to prevent them from being synced or shown in search results. So this is again imperfect and I would highly suggest that you don't use this for sensitive use cases even if you have a team of 10 or less. If you take a peek at the settings here and scroll to the very bottom, you'll see everything that's grayed out here is for the enterprise plan. And you can see with the enterprise plan, you also get connected to Google Drive, One Drive, and even beta intelligence for Apple. And whether or not you would be the best pick for enterprise really depends on the number of seats you have. So the recommended number of seats for enterprise is 150 users. So one, that is expensive at scale, and two, this would have to make real sense from an operational standpoint to be worth it. Unlike the chatbt experience, with enterprise, it's different to set up the actual credentials. So you'd actually have to create a service account on Google cloud which they have full instructions for and then use that to be able to communicate with the Google Drive and then provision exactly who should have access to what and to what extent. If you're looking to set this up, OpenAI has an internal FAQ document on setting up the admin manage setup for enterprise and it walks you through step by step how to set up the service account on Google Cloud, set up a new project. And there's quite a few steps they have to follow, but they go through it literally step by step here until it's ready to go. And by the end of it, you have a private key as well as a workspace that's fully set up that you can then sync the credential to CHVt to enable that provisioning. By the end of the Google Cloud setup, you'll be able to go back into GPT and then set up your service account key here and then put your admin email address for the service that was set up and then go through the wizard here, step one of five, and by the end of it, you should be able to sync, but now with the peace of mind that everything is synced exactly the way you like it. For those of you not dealing with super sensitive data in a pretty small team, looking into this feature of chatb teams might save you a lot of headaches of creating custom GPTs or creating knowledge bases or vector databases that you have to continually update over time and find ways to get it to speak to chat GBT. If you found this video helpful, let me know down in the comments below. And if you are a business owner, entrepreneur, or business leader and find tips and tricks like this super useful, consider joining my paid community called Early AI Adopters that you'll find in the second link in the description below. I'll see you next time.