.svg){kind=spacer}
Details
- Google DeepMind announced Deep Research and Deep Research Max, autonomous research agents powered by Gemini 3.1 Pro.
- Deep Research prioritizes speed and efficiency for interactive apps requiring quick responses.
- Deep Research Max takes extra time for exhaustive searching and reasoning, suited for background tasks and deep context gathering.
- Agents navigate the web and custom data like internal documents or financial info to produce professional-grade, fully cited reports.
- New features include arbitrary MCP support for secure connections to proprietary or third-party data, and native generation of presentation-ready visuals.
- Developers can start building via the Gemini API.
Impact
Google DeepMind's Deep Research agents advance autonomous AI research tools by integrating Gemini 3.1 Pro with MCP for secure custom data access, positioning Google ahead in agentic workflows. This pressures rivals like OpenAI's o1 series and Anthropic's Claude, which offer reasoning but lack comparable native MCP extensibility for enterprise data. By enabling cited reports and visuals directly via API, it lowers barriers for developers building production apps, potentially accelerating adoption in finance and research sectors amid growing demand for verifiable AI outputs.
