Summary

  • Google’s Deep Research tool in Gemini might soon allow users to upload their own documents and images as reference material.
  • This upcoming feature was spotted by @testingcatalognews and closely mirrors a core function of NotebookLM.
  • Unlike NotebookLM, however, Gemini’s Deep Research will likely combine uploaded files with real-time web browsing.

With the AI boom in full swing, every company has been racing to integrate AI into their existing products to make them better. While many have nailed the timing and seen their tools take off initially, the hype often fizzles out just as quickly. One of the few companies that has genuinely done AI right, surprisingly, is Google.

The tech giant has been consistently directing its efforts toward improving its AI offerings. Though I’m a tech journalist, I’m a full-time student by day. This is why, when I’m looking for ways to integrate AI into my workflow, it’s typically to cut down the amount of time I spend studying. Essentially, I look for ways to study smarter, not harder. Other than Google’s AI note-taking NotebookLM, Deep Research has become one of my go-to features. While a lot of AI tools now have a Deep Research feature, including ChatGPT, Gemini was the first tool to properly adopt it and rolled it out toward the end of 2024.

For those unaware, with Deep Research, all you need to do is type in the topic you’re researching, and Gemini will take it upon itself to scan the web for you and prepare a comprehensive report with key findings.

Tips to use Gemini in Google Docs

In addition to the content of the report, Gemini will also include all the sources it used to generate the report under a heading Sources used in the report, and the sources it explored but didn’t use under another heading Sources read but not used in the report. You can also ask Gemini questions about its report. For instance, I once found that the Deep Research report felt like a wall of text, so I asked it to write its main findings in bullets. Now, Deep Research might be getting another addition that’ll just make it all that much better, and it seems like it’s been inspired by NotebookLM.

You might soon be able to upload your own files to Gemini’s Deep Research

As spotted by @testingcatalognews via an X (formerly Twitter) post, Google is reportedly working on the possibility of uploading files to Gemini’s Deep Research tool.

This essentially means that you’ll be able to upload documents and images that Gemini can reference when researching and preparing its report. TestingCatalog uploaded an image to their X post that shows a pop-up saying Research with your own files, with a subtext Upload files for more tailored research insights.

Gemini Deep Research file upload feature

Image Credits: @testingcatalognews

I find this addition very NotebookLM-like. One of the primary reasons why the AI-powered note-taking tool is applauded by many is because it’s self-contained and only uses sources you’ve uploaded to answer your queries, generate AI-powered podcasts (which NotebookLM calls Audio Overviews), create mind maps, and more.

While this new Gemini addition will likely work differently, since it’ll also simultaneously browse the web, it seems like it’s combining NotebookLM’s best attribute with Deep Research’s real-time capabilities.