According to the first comprehensive benchmarking study published by LegalBenchmarks.ai in September 2025, AI tools in several cases produced contract drafts that were as accurate as, or more accurate than, those prepared by human lawyers. The study evaluated 13 AI tools across 30 real-world contract drafting tasks, where human lawyers achieved an average accuracy rate of 56.7%, while several AI systems scored higher than this benchmark.
The study assessed three dimensions of performance: output reliability, usefulness, and workflow integration through evaluation of 450 task outputs and a survey of 72 legal professionals. The top-performing AI tool, Gemini 2.5 Pro, achieved a 73.3% reliability rate, marginally outperforming the best human lawyer at 70%. In scenarios involving high legal risks, specialised legal AI tools raised explicit risk warnings in 83% of outputs compared to 55% for general-purpose tools, whilst human lawyers provided no such warnings. Legal AI platforms including GC AI, Brackets, August, and SimpleDocs all scored above average, although general-purpose AI tools slightly outperformed specialised legal AI platforms on reliability metrics.
Human lawyers demonstrated advantages in tasks requiring commercial judgement and context management, while AI tools proved more consistent in routine drafting and produced outputs in seconds compared to the human average of nearly 13 minutes per task. In workflow integration, specialised legal AI platforms differentiated themselves significantly from general-purpose tools, with two-thirds of tested legal AI products integrating with Microsoft Word, where most contract drafting occurs. Among the 72 surveyed lawyers who use AI for legal work, 86% employ multiple tools rather than relying on a single product.
Sources:
1.
Bob Ambrogi
2.
artificiallawyer
3.
Legal AI Research Team
