Anthropic Unveils New AI Tools Causing Major Drop in Software Stocks

Anthropic’s introduction of a new legal automation capability sent a jolt through markets, triggering a sell-off in software, information, and knowledge-based services stocks as investors reassessed how quickly artificial intelligence (AI) could erode long-standing revenue models across the sector.

Legal AI Sparks Market Shock

On January 30, Anthropic revealed 11 open-source plugins for its Claude Cowork platform. Among them, a legal-focused tool drew particular attention for its ability to automate contract analysis, nondisclosure agreement screening, and compliance reviews. These are tasks traditionally handled by paralegals and junior lawyers, prompting concerns that automation could reduce demand for entry-level roles and reshape early-career pathways in law.

Markets responded swiftly. Shares of Thomson Reuters slid 18%, while Pearson fell 7% and LegalZoom dropped nearly 20%. The decline spilled into adjacent industries, including software, financial services, and asset management, contributing to an estimated $285 billion loss in combined market value.

Investors debate disruption versus evolution

Scott Dylan, founder of Nexatech Ventures, described the sell-off as not signaling that AI agents will immediately take over these businesses, but rather as investors beginning to factor in the longer-term risk that providers of foundational models could directly challenge traditional software platforms. He emphasized that this concern is realistic, not hypothetical.

Adding a broader perspective, Jonathan McMullan, an analyst at Schroders, said the market movement reflects a structural reassessment across the sector. As companies are able to achieve more with fewer staff due to technological advances, investors are increasingly questioning the predictability of earnings and challenging traditional per-user pricing models.

Investors are aggressively repricing these areas as the historical ‘visibility premium’ erodes; the speed of AI advancement makes long-term valuations harder to defend, particularly as AI tools allow businesses to do more with fewer staff, threatening the traditional model of charging per software user.

Jonathan McMullan

The turbulence reached beyond technology. Advertising groups Omnicom and Publicis declined 11.2% and 9%, respectively, while Australian cloud-accounting provider Xero recorded its worst session since 2013 with a 16% drop. Giuseppe Sersale, a fund manager at Anthilia, observed that AI is increasingly capable of performing the core operational tasks that support these business models, putting certain parts of the sector under ongoing strain.

AI’s Role in Law: Aid, Not Replacement

Within the legal profession, views are more nuanced. Joel Simon, founder and partner of Simon Perdue in Texas and New Mexico, emphasized that human judgment and credibility remain central. AI can assist with preparing and analyzing legal materials, but attorneys retain control over courtroom strategy, case presentation, and final judgment. Simon predicts that trial lawyers who integrate AI into their workflow will become more valuable over the next two to three years.

Dylan, however, expects a tougher adjustment. He believes clearly definable tasks are most likely to be automated, which could reduce certain human responsibilities. While the work itself may not disappear, traditional training pathways could become narrower.

Looking ahead, he added that humans will remain indispensable in roles requiring physical presence or close personal interaction, including healthcare, personal services, and skilled trades. Still, investors are already factoring in a challenging transition period.

Software Pricing and Workforce Transformation

Industry research points to a parallel shift in how software is sold, reflecting broader changes in pricing and delivery models

• IDC forecasts that by 2028, traditional per-seat software pricing will largely disappear as vendors adopt new approaches.
• Around 70% of software providers are expected to move toward pricing that charges based on actual usage, delivered outcomes, or the capabilities their software enables.
• Enterprise software companies are already testing hybrid strategies, with Bain & Company reporting that among more than 30 SaaS firms leveraging generative AI, roughly 35% have increased per-seat prices by including AI features, while another 35% combine usage-based and bundled pricing models.

On the employment front, a study from MIT estimates that current AI systems could already perform the tasks of 11.7% of U.S. jobs. Meanwhile, a 2025 report from the World Economic Forum suggests nearly 60% of workers will need to learn new skills to remain competitive in an AI-driven labor market. As AI becomes more integrated, workers will need to adapt, though human expertise will remain crucial for roles that require creativity, judgment, and direct interaction.