Automation moves from tasks to agents
Harvey reached a $15.6 billion valuation with annual recurring revenue exceeding $400 million. A&O Shearman uses agentic AI agents to automate complex legal workflows. These agents complete tasks in minutes that used to take several hours. The platform includes 597 pre-built agents to manage processes like due diligence and compliance reviews. Winston Weinberg, the CEO of Harvey, says the technology acts as the system through which legal work gets done. The deployment at A&O Shearman targets multi-step, long-horizon legal workflows across practice areas. These agents handle tasks like M&A, due diligence, and contract drafting. The company serves over 100,000 lawyers across 1,300 organizations in 60 countries. A&O Shearman became the first firm to deploy generative AI enterprise-wide in 2022. In 2023, the firm collaborated with Harvey and Microsoft to build ContractMatrix, an AI-powered contract tool. Harvey received seed funding from the OpenAI Startup Fund. The company recently raised $550 million to scale its AI agents. This funding accounts for 16% of the $3.35 billion invested in legal technology globally during 2026. The platform works with more than 25,000 custom AI agents to change how work is delivered.
Workflow depth versus research accuracy
Harvey provides deep workflow customization for due diligence, litigation, and transactional work. The deployment of the Vault product allows a user to load up to 100,000 files and create structured review tables spanning 10,000 files and 500 columns with per-cell verification and conditional logic for every single document in a set. This tool pulls up to 50 data fields per document with a 97% accuracy rate for term extraction. The Knowledge module provides research for legal, regulatory, and tax questions across 100 integrated legal data sources in 90 jurisdictions. The platform integrates with Microsoft Word, SharePoint, and Microsoft 365 Copilot to reduce platform switching for attorneys who spend most of their working hours inside these specific document environments.
On an independent 50-point benchmark, Harvey scored 38/50 on generic legal tasks. CoCounsel remains the primary choice for research-heavy litigation practices because it uses Westlaw and Practical Law databases. It provides a citation trail that helps lawyers verify legal authority. Legora, renamed from Leya in February 2025, serves as an alternative for firms in multiple European jurisdictions because it handles multilingual, cross-border documents. Legora carries a $5.5 billion valuation. CoCounsel plans run from $104 to $639 per user per month depending on the tier. LexisNexis+AI is another competitor that performs well on legal research accuracy.
| Category | Harvey AI |
|---|---|
| Reported monthly per-seat rate | $1,000 to $1,200 |
| Reported seat minimum | 20 to 25 |
| Publicly published pricing | None |
| Trial availability | None |
Harvey produces incorrect citations or fabricated case law occasionally. This hallucination risk means lawyers must perform manual fact-checking to avoid professional liability. Harvey maintains SOC 2 Type II, ISO 27001, GDPR, and CCPA certifications to protect client data.
The specialized competition
The market contains specialized tools for specific needs. Spellbook operates as a Microsoft Word add-in for drafting and redlining contracts. It functions differently than Harvey because it lives directly inside the document a lawyer is already editing. Spellbook scored 34/50 on contract drafting in independent testing. AILexSys provides an on-premise deployment option for firms in Hong Kong, Singapore, and the UK. This option helps firms with data sovereignty requirements. You probably already know that a single platform cannot solve every legal problem.
Firms use different paradigms based on their tasks. Research tools like Harvey and CoCounsel answer questions. Extraction tools like Kira Systems use machine learning to identify provisions. Purpose-built extraction infrastructure focuses on structured data for M&A due diligence. Large firms often use a hybrid strategy. They use commercial platforms for commodity tasks and build bespoke AI for differentiating workflows to maintain a competitive advantage. For M&A due diligence, Luminance and Ironclad focus on contract intelligence and lifecycle management. Can Harvey maintain its lead as OpenAI releases more specialized models for legal research? Harvey is the best choice for large firms needing custom automation and high-volume document review.
