Google’s NotebookLM is a Personalized Lite Language Model

Google personalized AI assistant NotebookLM is an experimental product that has been in early access since July. Now the company is integrating its new Gemini Pro LLM with NotebookLM and making it available to U.S. residents 18 and older. NotebookLM is engineered “to help you do your best thinking,” Google says, with documents uploaded to the service making it “an instant expert in the information you need,” allowing it to answer questions about your data. Unlike generic chatbots, NotebookLM draws responses from the documents you feed it, meaning it will be hyper-focused — a lite version of a custom trained model.

“NotebookLM now has new tools designed to help users organize their curated notes into structured writing projects,” TechCrunch reports, explaining that “you can select a set of notes and ask NotebookLM to create something new, such as a script outline, email newsletter or a draft of a marketing plan.”

NotebookLM can also suggest actions based on a user’s currently activities, offering summaries, or exposition, on text of what you’re reading. On a current writing project, it will offer suggestions based on the sources you’ve included in what you’ve written thus far.

One new feature Google says was based on requests from test users is “the ability to save interesting exchanges with NotebookLM as notes.” A new noteboard space “can easily pin quotes from the chat, excerpts from your sources or your own written notes.”

As it did previously, NotebookLM will automatically cite sources when it answers a question, “but now you can quickly jump from a citation to the source, letting you see the quote in its original context,” Google writes in a news release.

VentureBeat calls the inability to browse the Web a “major limitation that seems like it will hamper NotebookLM’s usefulness,” and points out this “seems like a major omission from a company whose biggest strength to date has been its crawling and indexing of the entire World Wide Web.”

“Google‘s entries into the generative AI space so far have yielded mixed results, but that isn’t stopping the search giant — owned by parent company Alphabet — from shipping new products, services and features,” VentureBeat says, noting that Gemini Pro is the middle sibling among three new AI models Google just announced.

The “Gemini Ultra larger context and higher parameter count (connections between artificial neurons) is due out next year, while a Gemini Nano is a smaller parameter version requiring less computing power designed for smartphones running Google’s Android open source operating system,” VentureBeat writes.

No Comments Yet

You can be the first to comment!

Leave a comment

You must be logged in to post a comment.