Crosby is using AI to democratize legal services for individuals and small businesses
Oct 8, 2025 with Ryan Daniels
Key Points
- Crosby closes $20 million Series A led by Index Ventures, positioning itself as the only AI-native law firm building internal orchestration rather than selling software to existing firms.
- Contract processing velocity accelerated from 1,000 agreements in 170 days to 1,000 every three weeks, with clients including Polymarket, Cursor, and Clay.
- The company routes work between human attorneys and AI models based on complexity, updating its routing logic every three to four weeks as model capabilities improve.
Summary
Crosby, an AI-native hybrid law firm, has closed a $20 million Series A led by Index Ventures, with participation from Lior Gil. The company is run by Ryan (last name not given in transcript) and positions itself as the only player in legal AI building the firm itself, rather than selling software to existing firms.
The traction numbers tell a clean story. It took Crosby roughly 170 days to process its first 1,000 contracts. It now processes 1,000 contracts every three weeks, with volume still accelerating. Clients include fast-growing tech companies Polymarket, Cursor, and Clay.
How it works
Crosby operates a licensed law firm staffed by human attorneys working alongside AI tooling. The core operational challenge is orchestration — deciding which contract terms and provisions route to a senior lawyer, a junior lawyer, or an AI model entirely. That routing logic is updated every three to four weeks as model capability improves, allowing progressively more complex work to shift toward automation.
For clients that already have in-house legal teams, Crosby positions itself as a second set of arms handling the backlog of non-strategic agreements — MSAs, NDAs, DPAs, sales contracts — that are low priority for senior counsel but create friction for sales and procurement teams.
Why a firm, not a software product
The investor pushback Ryan describes is the obvious one: why not sell this as software? His answer is that selling a copilot to lawyers is the limited bet. If models continue to improve to the point where they can handle most legal work, the company that has been continuously running and refining its own orchestration layer — with lawyers and engineers sitting desk-by-desk, building specialized models in-house — will have a structural advantage that a pure software vendor cannot replicate. Selling software to other firms means that tight feedback loop never forms.
The expansion story is early but directional. Crosby started in the Silicon Valley AI ecosystem, where clients' growth intensity pushed the team to deliver faster and more accurate contract review. Larger companies that were skeptical of an AI law firm a year ago are now approaching Crosby inbound, having watched it work for companies like Cursor that have since grown well beyond startup scale.