RSGI Reports Harvey Adoption Jump Among Legal Teams

RSGI's second report, based on research from April to June 2026 with 87 respondents across 60 firms and 27 in-house teams, finds that 68% of law firms and in-house teams are deploying Harvey-based AI agents, with power users saving an average of 11 hours per week. The study indicates a shift from pilot projects to broader operational use in legal workflows, driven by integration with document systems and Microsoft 365.

RSGI Reports Harvey Adoption Jump Among Legal Teams According to legaltechnology.com's coverage of RSGI's new report, RSGI released its second study, "The Accelerating Impact of Legal AI: Harvey as Foundational," based on research conducted April to June 2026 with 87 unique respondents across 60 firms and 27 in-house teams. The report, per legaltechnology.com, finds 68% of law firms and in-house teams deploying Harvey-based AI agents, 21% of law firms running more than 50 agents in production, power users saving an average of 11 hours per week up from 8.5 , and a rise in organisations with 30%+ power users from 34% to 50% . Editorial analysis: These metrics indicate a shift from pilot projects to broader operational use in legal workflows, with integration to document systems and Microsoft 365 cited as key enablers. What happened According to legaltechnology.com's coverage of RSGI's research, RSGI published its second report, "The Accelerating Impact of Legal AI: Harvey as Foundational," drawing on research conducted between April and June 2026 with 87 unique respondents across 60 firms and 27 in-house teams. The study was commissioned by Harvey and carried out independently by RSGI, per legaltechnology.com. Key findings, per legaltechnology.com - • 68% of law firms and in-house teams are deploying Harvey-based AI agents, and 21% of law firms report running more than 50 agents in production. - • 89% of law firms report they can take on more work because of Harvey, per the RSGI data. - •Power users at law firms now save an average of 11 hours per week, up from 8.5 hours six months earlier. - •Among firms that can track outcomes, 59% report increased lawyer utilisation, 44% report increased revenue, and 53% report increased profitability for lawyers using Harvey. - •The share of organisations with 30% or more power users rose from 34% to 50% in six months, and 57% of law firms say a lawyer can become a power user in under three months. - •Over half of respondents describe Harvey as a foundational technology for legal services delivery 55% of law firms, 48% of in-house teams . Editorial analysis These reported metrics align with patterns seen in other enterprise AI rollouts where rapid integration with core systems drives a shift from experimentation to production. Industry-pattern observations suggest that measured time savings and ability to handle additional work are common early ROI signals that accelerate broader deployment. Context and significance For legal practitioners and technologists, the report's specific operational metrics, weekly hours saved, increases in power-user penetration, and production-scale agent counts, provide concrete benchmarks for evaluating vendors and internal adoption. Industry context: Integration with document management systems, Microsoft 365 and MCP/API being described as "gamechangers" underscores the importance of workflow-level connectivity for useful automation. What to watch Indicators to follow include whether reported client discussions and mandates from in-house teams translate into documented fee or pricing changes, whether firms convert measured productivity into billing or capacity gains, and whether the share of power users continues to climb. Observers should also track independent replication of these metrics in other vendor or sector studies to validate the trend. Scoring Rationale The report provides concrete, production-level adoption metrics and ROI figures for legal AI, useful for practitioners benchmarking deployments. The story is notable within legal/vertical-AI contexts but not broadly industry-shifting. Practice with real Ad Tech data 90 SQL & Python problems · 15 industry datasets Active Search Campaigns by BudgetEasy /problems/sql/active-search-campaigns-by-budget High CPC Clicks & Poor Landing PagesMedium /problems/sql/high-cpc-clicks-poor-landing-page Campaign ROAS by Attribution ModelHard /problems/sql/campaign-roas-by-attribution-model 250 free problems · No credit card See all Ad Tech problems /problems/datasets/adtech