Large language models are useful chatbots such as ChatGPT and Microsoft Copilot, but as they become increasingly more popular, more and more potential uses are coming up for them. Google announced NotebookLM earlier this year, which promised an upgraded notetaking app with AI powers, and now it’s exiting a testing period and getting new features.
NotebookLM has officially launched in the United States, alongside the wider release, the AI-powered notetaking app is getting a few more features to make it more useful than ever. To start with, it’s based on Google’s new Gemini AI, which the company is trying to position as a competitor to the powerful GPT-4 model by OpenAI.
Another new addition is the “Noteboard.” It’s accessible from above the chat box, and this feature employs a card-based grid UI, allowing users to save NotebookLM responses, source excerpts, and personal notes. You can also now select multiple notes and instruct NotebookLM to perform various actions, such as summarizing, combining into a single note, creating outlines, research guides, or even transforming notes into different formats appreciate email newsletters or script outlines.
Another addition is that NotebookLM now able to dynamically suggest actions based on user activities. For example, when a user selects a passage from a source, NotebookLM automatically offers to summarize the text into a new note or assists in understanding technical language. Again, these suggested actions will really depend on what you’re doing at the moment, and they will show up sporadically in an attempt to help improve your workflow or, if it doesn’t need optimizing, just polish whatever you’re doing at the moment. It’s a pretty neat tool if you want to keep your workflow and sources as organized as possible.
If you want to check it out by yourself, make sure to visit Google’s NotebookLM landing page, adopt the terms, and give it a spin. Keep in mind that it could still produce the occasional inaccurate response, since this is all built on the same large language models that occasionally furnish false information in ChatGPT, Google Bard, and other chatbots. Many of the new features will also take some time to roll out.
Source: 9to5Google
