- Google has announced a major upgrade to its AI research tool, “NotebookLM,” featuring new capabilities exclusively for “Google AI Ultra” members, the company’s top-tier AI plan.
- The update introduces three major features: an “Agent capability,” improved reasoning, and expanded data export options.
- These advancements make the process of solving complex, multi-step research tasks easier than ever before.
On Monday, June 8, 2026, Google announced a significant upgrade to its AI research tool, “NotebookLM,” introducing new features specifically for members of its top-tier “Google AI Ultra” plan.
This major upgrade to “NotebookLM” for “Google AI Ultra” members includes three core features: the implementation of an “Agent capability,” enhanced reasoning performance, and support for a wide variety of data outputs. These features collectively simplify the process of tackling complex, multi-stage research tasks to an unprecedented degree.
Implementation of “Agent Capability”

Previously, “NotebookLM” required users to manually add source materials when creating a new “Notebook.” With the new “Agent capability,” the AI can now autonomously construct a “Notebook” even without pre-added sources.
For instance, simply by entering vague ideas or questions into the “Notebook” chat, the AI will autonomously search for and add relevant sources. Users can now conduct research through “NotebookLM” without needing to hunt for source material themselves.
Improved Reasoning Capabilities
“NotebookLM” chat has been updated with the latest Google AI model, “Gemini 3.5,” and the next-generation AI development tool, “Google Antigravity,” resulting in significantly improved reasoning capabilities.
The integration of “Gemini 3.5” and “Google Antigravity” not only provides more accurate and reliable information but also allows users to clearly track the AI’s thought process as it works. Furthermore, a secure cloud computing environment is now standard in the background of every “Notebook,” enabling users to write and execute code directly within the interface for deeper research and complex data analysis.
Additionally, over 100 curated software skills have been integrated, offering diverse new approaches for deepening the understanding of sources contained within a “Notebook.”
Expanded Output Formats

The “Studio” content generation feature in “NotebookLM” has expanded its supported output formats, allowing users to generate and download various file types based on the integrated sources within their “Notebook.”
Users can also provide detailed instructions during the file generation process, offering greater customization than before.
| ■ “NotebookLM” Expanded Output Formats |
| ・Data Visualization / Graphs: png / svg ・Documents: PDF / docx / Markdown / Text ・AI Image Generation: png / jpg / gif ・Structured Data: CSV / JSON ・Tables: Microsoft Excel (xlsx) ・Slides: Microsoft PowerPoint (pptx) |
Source: Google (https://blog.google/innovation-and-ai/products/notebooklm/better-research-notebooklm/)





