Legora reviewed 41 documents in minutes with GPT-6 Astra

OpenAI's GPT-6 Astra reviewed 41 complex financial documents in minutes, detecting all four planted errors with 100% accuracy and improving task performance by approximately 40% on a financial-review workflow. The demonstration suggests that AI can handle multi-document error detection at scale, potentially shifting human reviewers to exception handling in high-stakes document pipelines.

OpenAI https://openai.com/index/legora-financial-statement-review-with-astra Legora reviewed 41 documents in minutes with GPT-6 Astra Which summary reads better? Pick one — models revealed after.Both summaries are AI-generated. A frontier model caught all four planted errors across 41 documents in minutes while lifting task performance ~40% on a financial-review workflow, showing reliable multi-document error detection at scale rather than sampled spot-checks. If those recall numbers hold on your data, you can shift human reviewers from first-pass reading to exception handling on high-stakes document pipelines—but validate the zero-miss claim on your own planted-error benchmark before trusting it without a human backstop. GPT-6 Astra processed 41 complex documents in minutes with 100% error detection, cutting financial-review workflow time by ~40%. This means you can now ship high-stakes document pipelines that previously required manual review, slashing latency and cost while maintaining audit-grade accuracy—critical for compliance-heavy production systems.