Anthropic just came for Big Law — 20+ connectors, 12 practice plugins, Westlaw inside Claude

Anthropic just made its biggest vertical push to date. On May 12-13, 2026, the company released Claude for Legal — a bundle that includes more than 20 Model Context Protocol connectors, 12 practice-area plugins, and a separate Thomson Reuters partnership that wires Claude into Westlaw's 1.9 billion documents and 1.4 billion KeyCite validity signals.

Read the connector list. It is not a sample. It is the legal-tech stack. Harvey. Solve Intelligence. DocuSign. Ironclad. Definely. iManage. NetDocuments. Relativity. Everlaw. Consilio. Box. Datasite. Twelve more on top. Anthropic is not partnering with one or two legal-tech vendors as a launch story. It is connecting Claude to virtually every tool a lawyer touches in a working day and letting one agent carry context across all of them.

The 12 practice-area plugins extend the same pattern into the depth dimension. Commercial Legal. Corporate Legal — including M&A due diligence and closing checklists. Employment Legal. Privacy Legal. Product Legal. Regulatory Legal. AI Governance Legal. IP Legal. Litigation Legal. Each plugin starts with what Anthropic calls a setup interview that learns the firm's playbooks, escalation chains, risk calibration, and house style. The model adapts to the firm before the firm uses the model.

The Thomson Reuters partnership is the sharpest specific. Claude now reasons across 1.9 billion Westlaw documents and 1.4 billion KeyCite validity signals with a patent-pending citation ledger that makes every source traceable in one click. For litigators, the cite-checking workflow that has historically required a paralegal afternoon collapses into seconds.

Why this matters: Anthropic is doing in legal what Palantir did in defense — embedded engineers, vertical depth, model + workflow + integration as one bill. Legal services in the US alone is a $400 billion industry. McKinsey, Accenture, and Deloitte have been talking about disrupting it for a decade. Anthropic is shipping the disruption this week.

The countervailing angle Fortune captured: hallucinations are still showing up in legal filings. Lawyers using AI badly are still getting sanctioned. The bet Anthropic is making — and the bet Big Law is making by adopting — is that the productivity gains from the new connectors outweigh the residual hallucination risk, and that the firms who learn the workflow first capture the operational advantage.

What to watch: whether OpenAI ships a competing Legal vertical package within 60 days, whether the American Bar Association issues fresh guidance on AI use in filings, and whether the AmLaw 100 firms publish disclosed adoption rates by end of Q3.