The USA Leaders
February 05, 2026
On February 3, 2026, global software stocks experienced a sharp decline as investors struggled with a new reality: artificial intelligence is no longer just a productivity enhancement; it’s becoming a replacement for entire job categories.
At the center of the sudden decline was the Anthropic new legal AI tool, released as part of its Claude Cowork platform. Within hours, software companies specializing in contract review, compliance management, and legal briefing faced steep sell-offs.
Anthropic’s New AI Plugins
Until this week, Claude was what most people thought of when they imagined AI assistants: a sophisticated chatbot that could write, summarize, research, and reason through problems. It was powerful, but fundamentally passive.
Anthropic’s new ‘plugins’ changed this. They bundle pre-built skills, direct integrations with professional tools and data, and quick-action commands, transforming Claude from an assistant you consult into an agent that executes workflows.
The legal plugin can review contracts, triage NDAs, manage compliance checks, and generate briefings without human intervention.
For law firms and legal tech companies, this posed an existential threat to their business models.
- RELX and Wolters Kluwer both fell 11%.
- Thomson Reuters, LegalZoom, and FactSet dropped over 10%.
- Experian and the London Stock Exchange Group declined as much as 7%.
In total, the market shed $285 billion in stocks across the software, financial services, and asset management sectors, a sell-off that unfolded in less than 48 hours.
Why Claude Cowork Scares Investors?
The new plugins let Claude act as an AI agent that actively executes legal workflows. The platform can review contracts in minutes instead of hours, identify compliance risks across hundreds of documents simultaneously, and generate legal briefings without human review.
A task that once required a paralegal to spend a full day on now takes seconds. For a legal tech industry built on charging for human time, this represents an existential threat. Investors did not panic because Claude is smart; they panicked because Claude can now do the work that generates revenue for thousands of software companies, giving to the rise of the question that people often do ask these days:
Is AI Taking Over Human Jobs?
Yes, and no.
- Customer service representatives face the highest automation risk at 80%, followed by data entry clerks.
- Paralegals and junior lawyers now face similar threats from tools like Cowork
- Young workers aged 20-30 in tech-exposed occupations have already seen unemployment rise nearly 3 percentage points since early 2025.
But the job picture is more complex.
In 2024, AI created approximately 119,900 direct jobs while only 12,700 jobs were lost to AI, a net gain of over 100,000 positions. While 85 million jobs will be displaced globally by 2025, 97 million new roles are expected to emerge, resulting in net job creation of 12 million positions. New roles like prompt engineers, AI trainers, and AI ethics specialists are emerging faster than traditional jobs disappear.
The Human Bottleneck in the Age of AI
The real tension isn’t whether jobs will continue to exist; they will. It’s whether the same people will still be doing them. A 55-year-old paralegal displaced by tools like Cowork is unlikely to retrain as a prompt engineer. A customer service worker in a small town won’t automatically find nearby AI-adjacent work.
While a Pew Research Center survey found that 52% of U.S. workers are worried about the future impact of AI in the workplace, these concerns often collide with reality. In practice, workers don’t get to vote on how automation is deployed.
The harder question is how quickly displacement accelerates, and whether policy, education, and labour markets can adjust fast enough to absorb the shock.
This is where Anthropic’s own rhetoric becomes complicated. While releasing tools that automate white-collar work, CEO Dario Amodei warns that humanity is entering “technological adolescence,” where AI advances faster than legal systems and society can manage.
If he’s right, then job retraining and policy adaptation may fall dangerously behind the pace of automation.
Related Developments in Anthropic AI and Legal Automation
The release of Anthropic new legal AI tool is part of a broader shift in Anthropic AI strategy from conversational assistance to full workflow execution. Recent Anthropic AI tool news suggests that legal automation is only the first vertical being targeted.
At its core, the platform combines advanced reasoning with enterprise integrations, positioning the Anthropic legal tool as a direct competitor to traditional legal software.
Unlike earlier systems focused on augmentation, the Anthropic AI legal tool performs autonomous AI contract review, compliance validation, and document triage at scale.
For corporate legal departments, this represents a move away from manual oversight toward automated risk detection.
AI compliance tools powered by Anthropic AI can now scan hundreds of agreements simultaneously, flag regulatory exposure, and generate standardized legal outputs with minimal human involvement.
As more organizations experiment with the Anthropic new legal AI tool, the distinction between legal professionals and legal software continues to blur—raising profound questions about pricing models, liability, and the future structure of legal work.
Conclusion
The market’s panic on February 3rd wasn’t irrational. It was a glimpse of a future that’s arriving faster than policy, education systems, or social safety nets can accommodate. The question now isn’t whether AI will transform work. It’s whether society will transform fast enough to manage that change without leaving millions behind.
Neha Shekhawat
