LIVE
OpenAI Partners with Chilean Ministry to Advance Public Sector AI Ambience Healthcare Expands Clinical AI Suite for Hospital Operations AI Models Vulnerable to Manipulation in Simulated Warfare Scenarios, Research Shows AWS Highlights Enterprise Governance and AI Workspace Quick Ahead of re:Invent 2026 LoanPro Integrates Amazon Bedrock to Automate Loan Servicing Workflows Google Brings Wallet App to iOS Users Across the United States Microsoft Reveals Cause of Global Outlook Outage AWS Outlines Phased Roadmap for Responsible Agentic AI Governance Google updates Gemini settings to separate chat history from AI model training Uber Eats and Zipline launch autonomous drone food delivery in Houston OpenAI Partners with Chilean Ministry to Advance Public Sector AI Ambience Healthcare Expands Clinical AI Suite for Hospital Operations AI Models Vulnerable to Manipulation in Simulated Warfare Scenarios, Research Shows AWS Highlights Enterprise Governance and AI Workspace Quick Ahead of re:Invent 2026 LoanPro Integrates Amazon Bedrock to Automate Loan Servicing Workflows Google Brings Wallet App to iOS Users Across the United States Microsoft Reveals Cause of Global Outlook Outage AWS Outlines Phased Roadmap for Responsible Agentic AI Governance Google updates Gemini settings to separate chat history from AI model training Uber Eats and Zipline launch autonomous drone food delivery in Houston

Uber Builds Legal AI Agent to Streamline Contract Reviews

🧠 AI Analysis 2 min read 👁 6

Uber's internal legal AI tool has cut average contract review time by over 20% using a mix of retrieval-augmented generation, deterministic rules, and historical lawyer feedback.

Uber Builds Legal AI Agent to Streamline Contract Reviews🧠AI Analysis

Photo: infoq.com

Key points

  • Uber's Legal Redlining Agent cuts review time by over 20% with 91% accuracy.
  • The tool operates inside Microsoft Word and combines RAG with deterministic rules.
  • Human lawyers retain final authority to review and approve all AI recommendations.

Ride-hailing giant Uber has deployed a Legal Redlining Agent that successfully reduces average contract review times by more than 20 percent while maintaining a 91 percent accuracy rate for its AI-generated decisions. Built as an add-in for Microsoft Word, the tool is designed to examine proposed contract modifications, evaluate legal risks, and propose responses aligned with corporate policy. Human lawyers maintain full responsibility for verifying and approving the system's output.

Development on the system began in early 2024, followed by an internal pilot with the company's legal department in 2025. The platform utilizes a combination of retrieval-augmented generation, past lawyer feedback, and a deterministic rules engine. While early iterations relied solely on legal playbooks, the team introduced a feedback loop to capture human reasoning and final text edits. This allows the system to pull relevant historical interactions using similarity search, metadata filtering, and a second-stage language model evaluation that prioritizes recent decisions.

The underlying technical architecture pairs a React and TypeScript frontend with a Python backend managed via LangGraph, alongside OpenSearch for vectorized rules and Redis for session management. A dedicated deterministic rules engine enforces mandatory legal requirements and overrides model output when necessary. Looking ahead, the engineering team is investigating agentic harnesses and legal knowledge graphs to expand workflow automation and semantic consistency.

Source infoq.com — infoq.com/news/2026/11/uber-legal-redlining-agent/?utm_campaign=inf… ↗

Contact the newsroom

Got a tip, a proposal or a partnership question? Write to us and we will reply shortly.