Spellbook Raises $50 Million Series B to Expand AI Contract Review Platform

Spellbook has secured a $50 million Series B round at a $350 million post-money valuation, led by Keith Rabois at Khosla Ventures with participation from Threshold Ventures, Inovia Capital, Bling Capital, Moxxie Ventures, Path Ventures, and Jean-Michel Lemieux. The financing brings total funding to over $80 million and arrives as the company reports more than 10 million contracts reviewed on its platform, adoption across nearly 4,000 law firms and in-house teams in 80 countries, and a revenue trajectory on pace to triple this year.

“We’re at the spreadsheet moment for lawyers,” said Scott Stevenson, CEO and co-founder of Spellbook. “Just as spreadsheets transformed accounting, large language models are transforming law after 20 years of technological stagnation.”

Why This Round Matters Now

Transactional legal work has long been constrained by fragmented tools, version sprawl, and the grind of back-and-forth edits. The timing of this raise reflects a broader shift: legal teams are moving beyond experiments and into production use of AI that integrates with familiar workflows. The emphasis here is not on replacing attorneys but on accelerating the tasks they already perform—drafting, redlining, benchmarking, and coordinating across multiple documents—so that strategic judgment and negotiation get more of the calendar than administrative labor.

The company’s traction signals that AI-assisted contracting is crossing from pilot to standard practice. Volume figures—millions of contracts reviewed and thousands of teams onboarded—suggest the category is no longer a novelty. Just as spreadsheets once standardized financial modeling, contract-native AI is beginning to standardize how organizations approach risk, obligations, and deal speed.

Built for Where Lawyers Work

A defining product choice is meeting users inside Microsoft Word rather than asking them to adopt a new interface or migrate documents elsewhere. That “Cursor for contracts” ethos shows up in the workflow: attorneys remain in control of drafting and redlining while getting targeted suggestions, risk flags, and clause alternatives in the same document window. The result is less tool switching, fewer copy-paste errors, and faster iteration cycles with counterparties.

Spellbook is positioning its AI to be transparent and citeable during negotiations. Attorneys don’t just need a suggested clause; they need rationale they can justify to clients and opposing counsel. That is the bar for adoption in high-stakes settings, and it guides the product’s emphasis on market comparables, learned preferences, and deal history as a source of truth rather than generic boilerplate.

From Review to the Full Transactional Stack

The new capital is earmarked to push beyond review into the broader spectrum of transactional work. That includes deeper contract intelligence—surfacing negotiation positions grounded in real-time market patterns—along with features that adapt to each firm’s style guides, playbooks, and historical outcomes. Recent releases point in this direction: Market Comparison helps teams benchmark language against what’s common in similar agreements, while Preference Learning tunes outputs to a firm’s or department’s typical risk posture so drafts come out closer to “house style” on the first pass.

Another pillar is multi-document orchestration. Complex matters rarely live in a single file—they involve term sheets, master agreements, schedules, and side letters that must align. By coordinating drafting and checks across these related artifacts, AI can reduce costly inconsistencies and late-stage surprises that otherwise surface during diligence or after signature.

Product Direction: Data-Grounded, Preference-Aware, Agent-Capable

Looking ahead, the roadmap converges on three themes:

• Data-grounded suggestions. Negotiations improve when proposed edits are backed by comparable language from similar agreements and current market tendencies. Expect more instrumentation that turns each round of edits into structured signals—informing playbooks, default fallbacks, and negotiation strategies over time.

• Firm-specific learning. Preference Learning and adjacent features allow the system to internalize a team’s positions on indemnity scope, liability caps, confidentiality carve-outs, and more. The closer first drafts are to “house standard,” the less time is lost in repetitive redlining.

• Multi-document agents. Spellbook Associate targets the reality that transactions span families of documents. An AI that plans, drafts, checks, and revises across those files—under attorney oversight—can eliminate mismatches and reduce the final miles of diligence risk.

The Road Ahead: From Faster Documents to Faster Deals

The most important implication of this raise isn’t merely faster contract review; it’s faster and more predictable dealmaking. As data-grounded drafting becomes common, counterparties will increasingly converge on standard positions earlier, reserving human-to-human time for the truly bespoke issues that define the economics of a deal. In the near term, think of AI as the force multiplier that converts institutional knowledge—those unwritten norms and partner preferences—into a living system that guides every matter, not just the ones a senior partner can personally touch. In the longer arc, multi-document agents coordinating entire closing sets will compress timelines, lower error rates, and shift legal capacity toward strategy and client counseling rather than document wrangling.

For legal teams navigating tighter budgets and higher expectations, that shift is transformative: less friction between business intent and signed agreement, and fewer bottlenecks when the organization needs to move. With this Series B, Spellbook is betting that the next phase of growth in transactional law will be won by tools that feel native, learn continuously, and turn the contract stack into a genuinely intelligent system—one that makes lawyers faster today and makes the business meaningfully faster tomorrow.