Anthropic targets legal sector with 12 new AI tools

Anthropic is escalating its push into the professional services market with 12 new legal AI tools, a move that aims to automate entire workflows and directly challenges incumbent software providers. The May 12 announcement details new plugins for its Claude Cowork platform that integrate with established legal software from DocuSign and Thomson Reuters, moving far beyond simple text generation to handle complex tasks from contract review to bar exam preparation.

"The interest from lawyers in using Claude has been significant," Mark Pike, Anthropic's associate general counsel, said in a statement. He noted that lawyers are the most active professional group using the platform, aside from software developers, with over 20,000 signing up for a single workshop on using Claude for legal work.

The new tools include a "commercial counsel" function for reviewing vendor agreements and specialized aids for law students. By connecting with essential industry software like Westlaw and even rival AI legal service Harvey, Anthropic is embedding its technology directly into the daily processes of law firms. This follows a February tool release that contributed to a selloff in European software stocks, with the iShares Expanded Tech-Software Sector ETF (IGV) falling 1.38% on the day of the new announcement.

For investors, the strategy strengthens Anthropic's commercialization narrative ahead of a potential IPO, helping to justify a valuation that has reportedly been eyed at over $900 billion. The launch confirms a seismic shift where foundation models are bypassing intermediaries to target professional users directly, placing significant pressure on the $1 trillion enterprise software market.

From February Buzz to May Expansion

The May 12 launch builds on momentum from a quieter release in February 2026, when Anthropic first introduced legal plugins for Cowork. That initial move was framed as a simple extension for specialized business functions but sent ripples through the market, causing a selloff in shares of traditional legal software companies. Investors appeared to recognize Anthropic's strategy: positioning its Claude model not as a direct competitor to giants like Thomson Reuters, but as a more flexible API layer that enterprises could build upon.

Today’s expansion supersizes that ambition. The dozen new plugins represent a shift from off-the-rack AI tools to custom-tailored agents that are "matter-aware"—they understand the context of a case, track its stage, and suggest next steps. This capability to handle "large chunks of work" autonomously is what separates it from first-generation AI drafting tools and poses a more fundamental threat to existing legal tech platforms.

The Dual-Edged Sword of AI Capability

While Anthropic markets its tools for professional productivity, the power of its underlying technology carries inherent risks. A month ago, the company revealed a new model named Mythos, which it described as so "strikingly capable" at hacking and cybersecurity tasks that its release had to be restricted to a small group of trusted partners. This highlights the dual-use nature of advanced AI, a concern echoed by a recent Google announcement that it had disrupted a criminal group using an unnamed AI model to discover a zero-day exploit.

The incident, which Google confirmed on Monday, marks a moment cybersecurity experts have long warned of: AI being used to supercharge a hacker's ability to find and exploit unknown software vulnerabilities. While there is no evidence linking the attack to Anthropic's models, it underscores the security challenges facing the entire industry. As AI companies race for performance, they are also creating tools that could, in the wrong hands, automate the discovery of security flaws on a massive scale, potentially making the digital world more dangerous. The challenge for firms like Anthropic is to capture the immense commercial opportunity in professional services while responsibly managing the powerful, and potentially hazardous, capabilities they are creating.

This article is for informational purposes only and does not constitute investment advice.