C

CMS

CMS Scales AI to 7,200 Lawyers Globally, Hits 95% Adoption

Curated & reviewed by Peter Korpak, Founder & Chief Analyst, 100SignalsHow we verify
95% (initial rollout phase)Lawyer Adoption Rate
7,200Lawyers Scaled Firmwide
3,000+Daily Active Users (Summer 2025)

Vendor-reported figures — source: theapplied.co

The Challenge

CMS, one of the world's largest international law firms with 7,200 lawyers and 10,000+ professionals across 50+ countries, needed to deploy generative AI at global scale while meeting strict data governance and client confidentiality requirements. The challenge was rolling out AI across a highly distributed workforce spanning dozens of jurisdictions and practice groups, where any misstep could carry reputational and legal risk.

The Solution

CMS selected Harvey and ran a structured 12-month pilot with 300 lawyers, using Harvey's Assistant (document analysis, drafting, Q&A), Vault (secure bulk document analysis), Knowledge (legal and regulatory research), and Workflow Agents (automating repeatable legal tasks). Strong pilot results led the firm to commit to a full firmwide rollout to all 7,200 lawyers.

Results

Adoption hit 95% in the initial rollout phase, and by summer 2025 more than 3,000 lawyers were actively using Harvey daily. Lawyers are reviewing larger document sets faster and taking on types of work that were previously impractical, freeing up time for higher-value activities.

Key Takeaways

  • A phased pilot-to-scale approach (300 lawyers in the initial pilot phase, then firmwide to 7,200) de-risked one of the legal sector's largest GenAI rollouts.
  • Bundling multiple capabilities (assistant, secure bulk document review, research, workflow automation) addressed varied legal workflows rather than a single use case.
  • Data governance and client confidentiality were treated as prerequisites, not afterthoughts, for global AI deployment at a law firm.

Share:

Details

Industry
Law Firms
Company Size
Enterprise
Company
CMS
Quality
Curated
Source published
Jul 1, 2025
Last verified
Jul 28, 2026

Have a similar implementation?

Share your customer's AI results and link it to your vendor profile.

Submit a case study →