AI Playground: Microsoft Copilot & Google NotebookLM

[00:00:00] Tessa Burg: Hello and welcome to another episode of Leader Generation brought to you by Mod Op. I’m your host Tessa Burg, CTO here at Mod Op. And today, and for at least once a month, I’d like to invite you into our AI Playground. We’ve been talking to our clients this past year and we’re getting a lot of questions about what are the best AI apps to use?

[00:00:27] Tessa Burg: How do I make sure that I have the right talent and skills within my organization to get the most value out of our data and our technology stack? And of course, What happens with my staff and my team as AI becomes more prevalent. So what we’re going to do is invite more experts from Mod Op to the table to give you their perspective and their actual use cases and applications on apps that they’re testing welcome inside the AI Playground.

[00:00:56] Tessa Burg: You’re going to love it. It’s going to help you navigate. How can you not only pick the right tech for your business? How can you leverage your data to get the most value and differentiate from competitors and provide your staff a safe, secure environment that protects your brand and helps them accelerate their career

[00:01:14] Alyssa Curci: Hi there. My name is Alyssa Curci, and I am the SVP of Customer Delivery within Mod Op in our strategic consulting group. The application that I tested is Copilot for Microsoft product. And specifically, the problem I’ve been trying to solve is that it’s very challenging, particularly when on a meeting, especially with clients or even internally, where I’m sharing my screen, sharing some content.

[00:01:41] Alyssa Curci: And then, in addition, I’m trying to capture meeting minutes, notes, action items, follow ups. Quite regularly, I’ll have three screens and a piece of paper and I’ll be jotting things down while I’m trying to toggle between different applications. So what Copilot has really helped me do is capture action items and meeting minutes and really just help with that administrative function for me while I’m in a meeting.

[00:02:09] Alyssa Curci: And what’s great about it is it’s embedded within the Microsoft suite of products, so it’s already in Teams, it’s already in my Outlook email, and I really am just able to click a button and leverage it.

[00:02:22] Alyssa Curci: So now I’ll open Copilot and I’ll say, I can ask Copilot anything. There’s also prompts that they give you, which is recap the meeting so far, list action items, suggest follow up questions, what questions are unresolved.

[00:02:35] Alyssa Curci: I’ll just go to recap meeting so far, which I think will be good. I can still go about presenting content on the screen. I can still conduct the meeting and really Copilot comes up with everything for me talks about the key topics . And then if we were in a meeting where There were going to be a lot of action items. We could also ask for the action items as well as perspectives by topic and a handful of other capabilities.

[00:03:00] Alyssa Curci: Another neat functionality of this is it’s really easy to copy and paste. I can just copy it and then I can go ahead and paste it somewhere else. So, for example, in a Word document, and as usual, it gives this little alert message, AI-generated content, be sure to check for accuracy.

[00:03:18] Alyssa Curci: So, this doesn’t replace the functionality of a human in sending out meeting minutes and action items. I always review anything that Copilot has developed for me, but for the most part, it’s spot on, and other than changing a few spellings of people’s names and whatnot, it generally is very, very accurate.

[00:03:39] Alyssa Curci: I would say that on a scale of 1 to 5, with 5 being the highly recommend, I would probably give this a 4 out of 5. I don’t know that, it is doing something for me that I couldn’t do myself. It’s really just saving me time.

[00:04:00] Alyssa Curci: I will say that Copilot has other functionality. You can just paste anything into the email. Draft with CoPilot. Coaching by CoPilot. So coaching can analyze the email and give you some recommendations. Be more professional. The email has a straightforward tone. It could come across as a bit abrupt and demanding. So, to be very honest, generally, I want a straightforward tone, definitely don’t want to come across as abrupt and demanding, but it is definitely good to have this suggestion, you know, start with a more polite tone. Use complete sentences. But the other thing that’s really helpful is I didn’t have complete sentences everywhere. You can draft with Copilot. You can do auto rewrite, make shorter, make longer, change tone, add more detail, make more creative. So I’ll just do auto rewrite for an example. It generated a little bit more of content for me.

[00:05:01] Alyssa Curci: That being said, you can always keep the Copilot suggestion, discard the changes, go back. So I’ll discard. But you get the idea. I think if you have a longer email with a lot of different items and you really want to make it succinct and short and have someone checking it over, it’s a really helpful tool.

[00:05:21] Alyssa Curci: There are other tools out there that do similar functions, but for me, when it comes to Copilot, the fact that it’s just already embedded in the app as I’m using it in real time makes it the easiest to use for me and why I pick it over others. Thank you, and I hope this is helpful.

[00:05:39] Jonathan Murray: This is Jonathan Murray. I’m Chief Strategy Officer at Mod Op and I work with the team at Mod Op Strategic Consulting. The application we tested was Notebook LM by Google. You can get to it by going to NotebookLM. Google. com. It’s a product that came out of one of their innovation processes and they partnered with an investigative journalist from the New York Times to create the initial product.

[00:06:05] Jonathan Murray: And it’s designed to essentially accept a bunch of source material and then to have you ask and interact with that material. So have you ask questions of it. It uses the Google’s Gemini model on the back end. Microsoft Mechanics Uh, but all of the answers are sort of constrained by the source material that you provide.

[00:06:23] Jonathan Murray: The problem or challenge that we used it for was an engagement with a large financial services client where we had hours and hours of interviews and we had taken the transcript of those interviews and normally we would go through the transcripts and manually , extract the insights from those and, the observations from the interviews and our recommendations.

[00:06:44] Jonathan Murray: And NotebookLM was literally able to cut that work process, by three quarters. , We were able to upload the transcripts from the interviews and then have NotebookLM do all of the summarization to derive insights, to answer specific questions that we had that we needed to provide guidance to the client on.

[00:07:02] Jonathan Murray: And it was a massive acceleration of the work that we did for that client. The features and aspects of the app that we really like are the fact that it’s privacy safe. So, unlike ChatGPT or any of the other large language models, you’re not uploading your source material and then that’s being, ingested into the back end of their model for reuse.

[00:07:25] Jonathan Murray: This model, when you upload your source material, it’s private to you. You can share your notebooks with other people, but other than that, the material stays where it is. It doesn’t get ingested by Google, and it doesn’t get reused, for the purposes of other customers. So it’s privacy safe, and it’s very interactive.

[00:07:41] Jonathan Murray: You know, it answers real questions. It’s very powerful LLM on the back end. And you can, get to , pretty complex answers pretty quickly. There aren’t really, and it’s one of the things about NotebookLM, there aren’t really, many competitors to this out in the market at the moment.

[00:07:56] Jonathan Murray: At least not ones that we’ve been exposed to. ChatGPT, you can upload documents and ask questions, but again, in the enterprise version, it’s safe, but on the public version of it, you’re obviously uploading your content, to the rest of the world as well. And it’s not as insightful as NotebookLM when you ask it questions about that content.

[00:08:17] Jonathan Murray: We now use NotebookLM as a standard part of our discovery process. So when we’re working with clients, we basically, make sure that every interview is recorded, that we have the transcripts from every interview, and those interviews get converted into PDFs.

[00:08:32] Jonathan Murray: And that’s one of the constraints of NotebookLM. It doesn’t It doesn’t accept a whole bunch of file formats. It just accepts PDFs and audio at the moment. I’m sure that will expand over time. But as long as you can get whatever content you have in PDF format, including PowerPoint presentations or Google Slides or whatever, it will be able to ingest them and it will be able to actually derive insight from those documents.

[00:08:53] Jonathan Murray: It’s a standard part of our operating procedure now because it saves us such a massive amount of time in terms of just crunching through all of that raw material and, building structured, summaries and insights, that essentially clients are paying us to deliver for them.

[00:09:10] Jonathan Murray: On a scale of one to five, it’s certainly up there. It’s close to a five. There’s obviously room for improvement with every product. We’d love to see it be able to do some additional things, but at the moment, compared to anything else that we have and the value that it’s creating, it’s a five for us.

[00:09:25] Jonathan Murray: So I hope that’s helpful. And I would encourage everybody to go take a look at Notebook LM and see how you might be able to use it in your work. Thank you very much.

[00:09:35] Tessa Burg: We’re looking forward to your feedback. Feel free to send in apps. And if you want to hear other episodes from Leader Generation, we’ll continue to build on our client community conversations with insight from everything on PR, marketing, web development, app development, and more. Thanks for listening. And I hope you enjoy.