Google has launched Gemini for legal work, a preview tool designed to automate contract review and legal research. The system connects via Model Context Protocol connectors to existing firm infrastructure including iManage, NetDocuments, DocuSign, Everlaw, RelativityOne, and Thomson Reuters HighQ. It also integrates with specialised AI services like Harvey to respect current access permissions. Partners such as Deloitte sell ready-made agents for tasks like contract summarisation. The product targets law firms and corporate legal departments alongside a similar offering for financial services, with healthcare and life sciences versions scheduled for release.
The actual significance lies in standardising how large models access restricted data sources. These industry-specific packages rely on pre-built prompts known as skills that agents can execute. The underlying AI models remain identical to standard products, meaning the value comes from connecting generic intelligence to specific databases rather than new technology. This approach allows firms to deploy agents without rewriting core workflows or exposing sensitive data. The strategy shifts the burden from building custom integrations to adopting pre-configured connectors.
- Integration relies on Model Context Protocol connectors
- Partners sell ready-made agents for specific tasks
- Underlying models are identical to standard products
