The legal AI market has matured fast. Two years ago, most law firms were still cautiously watching from the sidelines. Today, the conversation has shifted from “should we look at this?” to “which tool is actually right for us?” That's a better question, and it deserves a direct answer.
This overview covers the six legal AI tools that come up most consistently in firm evaluations: Harvey, Spellbook, Lexis+ AI, CoCounsel, Luminance, and Kira Systems. My goal is not to rank them in a league table. That misses the point. These tools serve fundamentally different use cases and different firm profiles. The right question isn't “which is best?” It's “which is best for your firm?”
A note on my perspective
I've sold CoCounsel, Westlaw, and Practical Law to attorneys and law firms through my time at Thomson Reuters. My perspective on legal AI comes from the sales side — understanding how firms evaluate and buy these tools — not from being a daily practitioner. For the tools I haven't sold directly (Harvey, Spellbook, LexisNexis, Luminance, Kira Systems), I'm drawing on what firms report, what's publicly documented, and the patterns I've observed in how legal teams think about software decisions. Where I'm speaking from direct experience, I'll say so.
How to think about choosing
Before looking at specific tools, it helps to be honest about your primary use case. Legal AI tools tend to cluster around three core workflows:
Most firms that end up with the wrong tool picked one in the wrong cluster. A litigation-heavy firm that buys Spellbook, or a corporate practice that signs up for a research-first platform, will underuse the tool and be frustrated by what it can't do. Use this as your first filter.
The tools, one by one
Harvey
Enterprise AI for complex legal work
Harvey is the name that comes up most often in conversations about serious legal AI at large firms. Built on Azure OpenAI infrastructure with a custom-trained model developed in partnership with OpenAI, Harvey has a strong reputation for research, brief writing, regulatory analysis, and complex legal drafting. The product suite includes an Assistant for research and drafting, Vault for high-volume document review, and a Workflow Builder for automating multi-step legal tasks. Harvey has Microsoft 365 integrations and partners with Intapp for ethical wall enforcement — making firm-wide AI use permissioned and auditable across client matters. The platform reached approximately an $8 billion valuation in a December 2025 funding round, reflecting substantial adoption at Am Law 100 firms and sophisticated in-house legal teams.
Best for
Large firms and in-house teams with complex, high-volume legal work
Not ideal for
Solo and small firms, or practices that primarily need contract drafting in Word
Spellbook
AI contract drafting inside Microsoft Word
Spellbook is the most accessible entry point into legal AI for most firms. Powered by GPT-4o, it works as an add-in inside Microsoft Word — attorneys draft as they normally would, and Spellbook reads the document and makes AI-powered suggestions in context. The platform benchmarks contract language against more than 2,000 industry standards, flags deviations from market norms, and offers Spellbook Associate, an agent product for more complex workflows. Used by more than 4,000 law firms and in-house teams, Spellbook operates a zero data retention policy — documents are never stored or used to train AI models — and is SOC 2 Type II, GDPR, and CCPA compliant. A 7-day free trial is available, making it the only major legal AI tool you can evaluate without a sales conversation.
Best for
Contract-focused attorneys who work primarily in Microsoft Word
Not ideal for
Firms whose primary need is legal research, litigation support, or large-scale due diligence
Lexis+ AI
AI-powered research from LexisNexis
Note: LexisNexis rebranded this product to 'Lexis+ with Protégé' in February 2026, reflecting expanded AI capabilities. The platform now runs on a combination of Anthropic's Claude and OpenAI models, offering both Protégé Legal AI (optimized for legal research and drafting) and Protégé General AI (for broader tasks). If your firm already uses LexisNexis for research, this remains the most natural AI upgrade — the database depth is substantial and the brand trust in legal is real. The tool integrates with Microsoft 365 via Lexis Create+. LexisNexis confirms it does not use customer data to train AI models and deletes uploaded documents after sessions — meaningful protections for client confidentiality.
Best for
Research-heavy practices already embedded in the LexisNexis ecosystem
Not ideal for
Firms not already on LexisNexis who want a standalone contract drafting or document review tool
CoCounsel
AI legal research and drafting from Thomson Reuters
This is the product I know most directly. At Thomson Reuters, I sold CoCounsel — alongside Westlaw and Practical Law — to attorneys and law firms. The integration with Westlaw is the core value proposition: CoCounsel can build a research plan, retrieve authority from Westlaw and Practical Law, search relevant documents, verify that citations are good law, and deliver structured work product within a single workflow. Since its launch in March 2023 as the first legal AI assistant built on GPT-4, Thomson Reuters has expanded CoCounsel significantly — it now draws on multiple frontier AI models including Anthropic's Claude, OpenAI's GPT, and Google's Gemini alongside proprietary Thomson Reuters technology. CoCounsel Legal, launched in 2025, adds agentic AI capabilities and deep research workflows. The platform has surpassed one million users across more than 100 countries. Thomson Reuters confirms customer data is not used to train third-party models.
I have direct sales experience with CoCounsel. My perspective here comes from conversations with attorneys and firms evaluating this product, not from being an end user myself.
Best for
Firms running on Westlaw who want AI built into their research workflow
Not ideal for
Firms not already using Westlaw who want a standalone tool
Luminance
AI document review built for complex transactions
Luminance takes a notably different approach to AI than most tools in this market: rather than building on a third-party model like GPT, Luminance has trained its own proprietary model on over 150 million verified legal documents, giving it the ability to recognize more than 1,000 specific legal concepts. The platform includes a traffic light risk system for rapid issue flagging, Luminance Autopilot for autonomous NDA negotiation (the AI can read a contract, remediate risks based on its training, and respond to counterparty changes), and integrations with Microsoft Word, Outlook, Salesforce, and virtual data rooms. Luminance holds ISO 27001 certification and confirms it does not use client data to train its models without permission. For firms doing high-volume M&A due diligence or complex contract portfolio analysis, the proprietary legal training is a genuine differentiator.
Best for
Firms doing high-volume M&A due diligence, document review, or contract analysis
Not ideal for
Firms whose primary need is research, brief writing, or contract drafting in Word
Kira Systems
Contract review and due diligence automation
Kira Systems has been in the contract review space longer than most, and Litera — which acquired Kira — launched a significant platform enhancement in January 2026: a hybrid AI architecture that combines generative AI with Kira's own proprietary models trained on over one million legal contracts, targeting 90% or better accuracy in contract analysis. The platform has more than 1,400 pre-trained clause types, a new Generative Smart Fields feature for creating custom extraction fields without prior training, and Concept Search for finding legal concepts across large document sets. Kira is used by 71% of Fortune 100 companies across M&A, real estate, banking, and IP matters. The Litera acquisition has accelerated investment in the product, connecting Kira with Litera's broader drafting and transaction ecosystem.
Best for
Large firms with high-volume contract review and structured due diligence workflows
Not ideal for
Firms whose primary need is research, drafting in Word, or legal AI beyond contract review
Quick reference: which tool fits which firm?
| If your firm primarily needs… | Consider |
|---|---|
| Legal research at scale | Harvey, CoCounsel, Lexis+ AI |
| Contract drafting in Microsoft Word | Spellbook |
| M&A due diligence and document review | Luminance, Kira Systems |
| AI built on top of Westlaw | CoCounsel |
| AI built on top of LexisNexis | Lexis+ AI |
| Fast onboarding, smaller budget | Spellbook |
| Enterprise-grade research and drafting | Harvey |
| High-volume contract provision extraction | Kira Systems, Luminance |
What I'd tell any firm evaluating legal AI right now
Start with your primary use case, not with the product name. The firms that struggle with legal AI adoption are usually the ones that bought a tool that was impressive in a demo but didn't map to how their attorneys actually work.
If you already pay for Westlaw, talk to Thomson Reuters about CoCounsel before you look elsewhere. The integration advantage is real, and the procurement is simpler if you're an existing customer. The same logic applies to LexisNexis. If your firm runs on Lexis, Lexis+ AI (rebranded to Lexis+ with Protégé in February 2026) is the natural first conversation.
If your practice is contract-heavy and your attorneys live in Word, Spellbook is worth its 7-day free trial before you invest in a larger platform. The onboarding is fast, the feedback loop is immediate, and you'll know quickly whether it delivers value. It's also the only tool on this list where you can validate fit before a procurement conversation.
And if you're doing significant document review or due diligence volume, Luminance and Kira Systems are both worth a serious look. Luminance's proprietary legal AI — trained on over 150 million verified legal documents — and its Autopilot feature for autonomous contract negotiation are the kinds of specialist capabilities you won't find in a general-purpose research tool. Kira's January 2026 hybrid AI enhancement has meaningfully strengthened its accuracy and flexibility for large-scale contract review.
One question worth adding to every vendor conversation: what happens to my data? Ask specifically whether your documents are used to train or improve the model, which AI providers' infrastructure processes your data, and how the platform handles conflicting client matters. The answers vary significantly across these tools and they matter for professional responsibility compliance.
