Harvey
Harvey · Legal AI for Law Firms and In-House Legal Teams
A purpose-built AI platform for law firms and enterprise legal departments. Harvey handles legal drafting, document analysis, due diligence, contract review, and workflow automation — grounded in vetted legal databases and firm-specific documents, not general training data.
Best for
Law firms and enterprise legal departments that need AI grounded in legal databases and firm documents, with enterprise-grade compliance controls and DMS integration
Not ideal for
Solo practitioners, small firms without enterprise budgets, or anyone needing a general-purpose AI assistant outside legal and professional services
Who it's for
Attorneys and legal professionals at law firms and enterprise in-house legal departments
Capabilities
- Legal drafting: generates documents, memos, briefs, and presentations grounded in firm templates and precedents
- Document vault: secure storage up to 100,000 files per vault with bulk key-term extraction and structured comparison across large document sets
- Legal research via 500+ vetted sources: LexisNexis (US case law), EUR-Lex (EU law), EDGAR (SEC filings), and jurisdiction-specific databases
- Pre-built and custom workflow agents for multi-step tasks like lease summaries, due diligence extraction, and regulatory tracking — no coding required
- Document management system sync: iManage, SharePoint, and Google Drive
- Microsoft Word add-in for drafting inside existing firm workflows
- Microsoft Outlook add-in for summarizing email threads and attachments in-inbox
- Citations exportable in firm-approved formats for professional accountability
- Mobile apps for iOS and Android
- API and MCP (Model Context Protocol) support for custom integrations
Limitations
- Enterprise-only access — no self-serve plan, no public pricing, no free tier; requires a demo and sales process
- Only built for legal and professional services — not useful for general business workflows outside that domain
- Data freshness for legal knowledge sources is not publicly documented; database update cadence is unspecified
- Vault capacity is capped at 100,000 files — behavior beyond this limit is not documented
- AI-generated legal content still requires attorney review before professional reliance
Use cases
- Drafting legal documents, memos, and briefs grounded in firm precedents and templates
- Running due diligence review and structured extraction across large contract or disclosure document sets
- Researching case law, statutes, and regulations across US, EU, and other jurisdictions with cited sources
- Building automated workflows for repeatable tasks like lease summaries or regulatory update monitoring
- Summarizing email threads and document attachments inside Microsoft Outlook without leaving the inbox
Our take
Harvey's core advantage is that it is purpose-built for legal workflows rather than being a general AI with a legal prompt. It connects directly into the tools attorneys already use — Word, Outlook, iManage, DMS systems — and grounds outputs in vetted legal databases with citable sources. That specificity matters for professional accountability. The trade-off is access: enterprise-only pricing with no self-serve option means smaller firms and solo practitioners cannot realistically evaluate it. If your organization can get access, the firm-template grounding, secure document vault, and workflow agent capabilities are genuinely differentiated from using Claude or ChatGPT with uploaded documents.
Who should use it
Legal professionals at AmLaw firms, large corporate in-house teams, and enterprise professional services organizations that need AI with legal database integration, firm template grounding, and enterprise compliance controls.
Who should skip it
Solo attorneys, small firms, or teams outside legal and professional services — the enterprise-only model and legal-specific positioning make it unsuitable for general business AI needs.
Strengths
- Grounded in 500+ vetted legal data sources (LexisNexis, EUR-Lex, EDGAR) — not reliant on general training data
- Native integration into legal workflows: Word, Outlook, iManage, and major DMS systems
- Secure document vault for bulk analysis across large contract and disclosure document sets
- Workflow agents buildable in natural language — no coding required
- Enterprise compliance: SOC 2 Type II, ISO 27001, GDPR, CCPA with regional data processing (US, EU, AU)
- No training on customer data by default — contractual guarantee
Weaknesses
- Enterprise-only access — no self-serve plan, no pricing transparency, requires sales engagement
- Limited to legal and professional services — not designed for general business AI needs
- Database update cadence for legal knowledge sources not publicly documented
- AI-generated legal content still requires attorney review before professional reliance
Harvey pricing
Enterprise
Custom — contact sales
- No public pricing; requires demo request
- Access to full platform: Assistant, Vault, Knowledge, Workflow Agents
- Regional deployment: US, EU/Switzerland, Australia
- SOC 2 Type II, ISO 27001, GDPR, CCPA compliance
- SAML SSO, audit logs, IP allowlisting
Note: Harvey does not publish pricing. Access requires requesting a demo. There is no free tier or self-serve signup.
Technical specs
Modalities
Text, Document analysis, Spreadsheet generation
Where Harvey excels
Due diligence across large document sets
The document vault supports bulk extraction of key terms and structured comparison across thousands of contracts — significantly faster than manual review and exportable as structured tables.
Legal research with cited sources
Harvey pulls from LexisNexis, EUR-Lex, EDGAR, and other vetted databases and exports citations in firm-approved formats — meeting the professional accountability requirement that general AI tools cannot.
Draft generation grounded in firm precedents
The Word add-in and Vault integration let attorneys draft against the firm's own templates and precedent library, rather than starting from generic AI-generated text that needs extensive re-working.
Automated regulatory monitoring
Custom workflow agents can track regulatory updates, extract relevant provisions, and surface changes on a schedule — a repeatable process that previously required manual monitoring.
Harvey vs. competitors
Harvey vs. ChatGPT / Claude
General AI assistants can draft legal text but lack integration with legal databases, DMS systems, or firm-specific templates. They also do not offer the enterprise compliance controls — SOC 2, ISO 27001, data residency, contractual no-training guarantees — that regulated industries require. Harvey is narrower but significantly more grounded for professional legal use.
Harvey vs. Thomson Reuters CoCounsel
Both are enterprise legal AI platforms with legal database integration. Harvey emphasizes its workflow agent capabilities, open API/MCP integrations, and multi-DMS compatibility (iManage, SharePoint, Google Drive). Thomson Reuters CoCounsel is tightly coupled to Westlaw. The choice typically comes down to existing legal database relationships and firm DMS infrastructure.
Harvey vs. Perplexity
Perplexity is useful for exploratory web research but is not designed for professional legal workflows — no DMS integration, no firm template grounding, no enterprise compliance controls, and no support for bulk document analysis across contract sets.
Frequently asked questions
Is Harvey free to use?
No. Harvey is an enterprise product with no free tier and no self-serve pricing. Access requires requesting a demo and going through a sales process.
What legal databases does Harvey integrate with?
Confirmed integrations include LexisNexis for US case law and statutes, EUR-Lex for EU law, EDGAR for SEC filings, Gyldendal Rettsdata for Norwegian law, and FromCounsel for legal content. Additional sources may be available depending on the deployment.
Does Harvey train its models on my firm's documents?
Harvey contractually guarantees it does not train models on customer data unless the firm explicitly opts into a bespoke model arrangement. This is a contractual commitment, not purely a technical constraint.
Why can't I use ChatGPT or Claude for legal work instead of Harvey?
You can use general AI assistants for some legal drafting tasks, but they lack three things Harvey provides: integration with vetted legal databases (LexisNexis, EUR-Lex, EDGAR) for source-backed research, native connection to DMS systems and firm templates, and the enterprise compliance controls (data residency, no model training on client data, SOC 2/ISO 27001 certifications) that professional legal environments require. For low-stakes drafting or general research, Claude or ChatGPT may be sufficient. For client-facing work, due diligence, or anything requiring citable sources and compliance guarantees, the gap matters.
How does Harvey compare to Thomson Reuters CoCounsel?
Both are enterprise legal AI platforms built for law firms and in-house teams. The key difference is database coupling: CoCounsel is built around Westlaw, while Harvey connects to multiple legal databases (LexisNexis, EUR-Lex, EDGAR, and others) and integrates with multiple DMS platforms. Harvey also supports custom workflow agents and an open API. The decision usually comes down to whether your firm is already on Westlaw versus a more diverse database setup.
Why does AI-generated legal content still require attorney review?
Harvey generates drafts and research summaries grounded in legal databases and firm documents, but AI outputs can mischaracterize case holdings, miss relevant jurisdiction-specific nuance, or apply precedents incorrectly. Professional legal accountability requires that a licensed attorney review and take responsibility for any work product before it is used in a client matter. Harvey is a drafting and research tool, not a replacement for legal judgment.
Does Harvey work inside Microsoft Word?
Yes. Harvey has a Microsoft Word add-in that lets attorneys draft documents using DMS content, Vault files, and firm templates without leaving Word. There is also a Microsoft Outlook add-in for email summarization.
Integrations & fit
Alternatives to consider
Anthropic
Developers and writers who need the most accurate code generation, precise instruction following, and the ability to reason across very large documents
FreemiumOpenAI
Knowledge workers who need one AI subscription covering writing, research, coding, image/video generation, and team collaboration
FreemiumPerplexity AI
Fact-checking, current events research, and cited answers
FreemiumResearchers and students who need to ask questions about their own document collections with cited answers
FreeAbout Harvey
Harvey is not a general AI assistant adapted for legal use — it is built specifically for lawyers. It integrates with legal databases including LexisNexis, EUR-Lex, and EDGAR for source-backed research, and with document management systems including iManage, SharePoint, and Google Drive for secure storage and bulk analysis. Attorneys use it to draft from firm templates and precedents, extract structured data across large document sets, and build custom workflow agents without coding. It ships a Microsoft Word add-in and Outlook add-in so lawyers can work inside tools they already use. Harvey is sold exclusively to enterprise clients through a demo-based sales process — there is no self-serve access, no free tier, and no public pricing.
Are you the founder? Claim this listing →