Big Law builds internal AI platforms
Kirkland & Ellis plans to spend $500 million on internal AI projects. The firm intends to create its own legal AI platform trained on the collective intelligence of its attorneys. Freshfields signed a bespoke deal with Anthropic to build customized tools integrated into firm systems. Freshfields receives early access to Anthropic’s latest models because they share knowledge with the provider. These moves challenge the dominance of specialized startups like Harvey and Legora. Harvey and Legora raised over $2 billion in venture capital funding collectively. These two companies serve over half of the top 100 largest US law firms. Sixty-two of the top 100 largest US law firms posted gross revenue of over $1 billion in 2025. Many of these profitable firms want to own their core business technology rather than empower vendors that could eventually replace them. Law firms recognize that AI enables work to be done faster. This shift drives firms toward value-based models instead of traditional billable-hour models. Ethan Batraski of Venrock GP says Claude Code and new AI infrastructure solutions make it easy to rebuild Harvey or Legora in short order. Min-Kyu Jung, founder and CEO of Ivo, says firms must decide if they want to own their core business or empower vendors that could eventually replace them. Kirkland & Ellis moves toward a business model that looks more like a technology company. Can Kirkland & Ellis successfully replicate the capabilities of specialized vendors with its $500 million investment?
Specialized agents dominate legal workflows
Legora grew rapidly, reaching 800 customers in more than 50 markets within three years of launch. The platform uses consumption-priced agent credits. Harvey provides Vault for mass document analysis of up to 100,000 documents. Harvey costs between $1,000 and $1,200 per lawyer per month with a minimum of 25 to 50 licenses. Users frequently point to high seat costs and minimum license commitments at Harvey as reasons to look elsewhere.
| Tool | Primary Use Case | Pricing Model |
|---|---|---|
| Harvey | Enterprise workflows | Quote-only |
| Legora | Research and drafting | Quote-only |
| Spellbook | Contract drafting in Word | Not published |
| Bind | In-house CLM | Published ($90/seat/month) |
| AI Lawyer | Small practice assistance | Subscription ($19.99/month) |
Legora’s Tabular Review turns hundreds of documents into structured, queryable tables. This tool competes on price against Harvey. Spellbook works as a Microsoft Word add-in for contract drafting and review in small to mid-sized firms. Bind provides AI-native contract lifecycle management for in-house legal teams. Bind handles drafting, redlining, and approvals from a single tool. The Bind Business tier costs $500 per month and includes 5 users. You should match the tool to your actual job, not to the category’s momentum. Ironclad targets large in-house departments with dedicated legal ops. Purchasing data from Vendr shows a median annual cost of $40,000 for Ironclad. CoCounsel provides AI legal assistance by combining generative AI with the Westlaw research ecosystem. Some users prefer CoCounsel because it grounds answers in the Westlaw legal corpus, which helps prevent inconsistencies. Luminance operates as a UK-founded platform for document analysis and large-scale document processing. AI Lawyer acts as a self-service assistant for individuals and small businesses. Plans for AI Lawyer start at $19.99 per month for individual users.
Regulation and niche litigation tools
Steno raised $49 million in Series C funding to expand its AI-driven litigation platform and court reporting services. Savano Capital Partners led this funding round with support from First Round Capital and The Legal Tech Fund. The Los Angeles-based company uses Transcript Genius to help lawyers review depositions and legal transcripts faster. The company plans to use the new capital to expand operations across the United States, improve its proprietary technology platform, and accelerate development of AI-powered tools for legal professionals across the country.
The EU AI Act transparency rules for generative AI became effective in August 2026, requiring providers to ensure that AI-generated content is identifiable and that humans receive clear information about their interaction with machines. Rules for high-risk AI use cases in the administration of justice will apply starting 2 December 2027. The AI Act also prohibits nine specific practices, including harmful AI-based manipulation and the use of AI for social scoring. The Commission also introduced rules for General-Purpose AI models that became effective in August 2025. High-risk AI systems must also follow strict obligations regarding risk assessment and cybersecurity.
