Uber's AI Lawyer Assistant Cuts Contract Grunt Work by 20%
Original headline: Uber’s Legal AI Agent Cuts Contract Review Time by 20% with Rules and RAG — infoq.com
More in Programming and developer news
Uber has developed an AI-powered Legal Redlining Agent that's integrated directly into Microsoft Word. It analyzes proposed changes to contracts, assesses legal risks, and recommends responses based on a mix of retrieval-augmented generation (RAG), historical feedback from Uber's own lawyers, and hard-coded deterministic rules. The company reports it has already reduced average contract review time by more than 20%, achieving 91% accuracy in its AI-generated decisions. Crucially, lawyers retain final approval and responsibility for all recommendations.
This shift from a pure RAG system to one incorporating a weighted feedback loop is the key engineering breakthrough. The AI now retrieves and learns from thousands of past lawyer decisions—how similar clauses were actually handled, including the final reasoning and edited text—using a similarity search and a language model relevance filter. A configurable decay weighting prioritizes recent decisions while keeping useful older examples. For developers and legal teams, this represents a tangible move from abstract 'AI assistance' to a production-ready, domain-expert-controlled workflow that reduces repetitive work without removing human oversight. The takeaway is that the most effective enterprise AI isn't about replacing experts, but about meticulously augmenting their judgment with their own institutional knowledge.
Read the original at infoq.com
Satirical commentary on real tech news, summarised with AI. Read the original source for the facts. How ZeroDaily works
