On January 30, Anthropic released 11 open-source plugins for its Claude Cowork platform. Five days later, $285 billion had evaporated from global software stocks in what traders at Jefferies quickly dubbed the “SaaSpocalypse.”
The trigger wasn’t a single product launch. It was a convergence: disappointing earnings across the software sector, accelerating AI capabilities, and a legal-focused plugin that made Wall Street realize AI wasn’t just helping software companies build faster products. It was positioning itself to replace those products entirely.
What Anthropic Actually Released
Claude Cowork launched on January 12 as what Anthropic describes as “Claude Code for the rest of your work” — a general-purpose AI agent designed for non-technical knowledge workers. It can read files, organize folders, create documents, and integrate with existing business tools.
The plugins, announced January 30, bundle skills, connectors, slash commands, and sub-agents so users can configure Cowork as a specialist for specific roles. All 11 are open-source. They cover sales, finance, marketing, data analysis, and — the one that lit the match — legal.
The legal plugin automates contract review using playbook-based approaches, handles NDA triage, runs compliance checks, generates redlines with negotiation guidance, and flags liability risks. Anthropic marketed it as an “AI junior lawyer.” The company emphasized it doesn’t provide legal advice and requires attorney review — but the market didn’t focus on that caveat.
The Carnage
The February 3 selloff hit legal and data companies first, then radiated outward:
Legal and data firms took the hardest hits:
- Thomson Reuters plunged 15.83% — its largest single-day drop ever
- LegalZoom cratered 19.68%
- RELX (owner of LexisNexis) fell 14%, followed by another 1.5% the next day
- Wolters Kluwer dropped 13% — its worst session in decades
- CS Disco sank 12%
- FactSet fell 10.51%
Enterprise software giants weren’t spared: Salesforce, Adobe, DocuSign, Workday, and ServiceNow all got caught in the downdraft. The Goldman Sachs software basket dropped 6% in a single session. The iShares Expanded Tech-Software ETF fell 4.6%, extending losses to a sixth consecutive day.
The contagion went global. In India, TCS dropped 6% and Infosys fell over 8%. London Stock Exchange Group fell 13%. Blue Owl lost 9.76%. The Nasdaq 100 slipped 1.6%.
By some estimates, the broader damage to software and services stocks approached $1 trillion when accounting for the multi-day selloff.
Why This Plugin Caused More Panic Than GPT-4
AI models have been getting smarter for years. Why did an open-source plugin for lawyers trigger a market rout?
Because it exposed a specific, existential threat to how software companies make money.
Most enterprise SaaS revenue depends on per-seat licensing. Ten lawyers need ten LexisNexis subscriptions. Fifty salespeople need fifty Salesforce licenses. The math is simple: more employees, more revenue.
Claude Cowork inverts that. One AI agent, properly configured with plugins, can do work that previously required multiple employees using multiple software tools. Thomas Shipp at LPL Financial summed up the fear: “Why do I need to pay for software…if internal development of these systems now takes developers less time with AI?”
This wasn’t the first time AI spooked markets. Nvidia lost roughly $600 billion in value in January 2025 when DeepSeek demonstrated cheap AI training — but that stock recovered in months. The difference here is that Claude Cowork doesn’t threaten the infrastructure layer. It threatens the application layer — every piece of software that sits between AI models and end users.
Only 71% of S&P 500 software companies beat revenue expectations this earnings season, compared to 85% across broader tech. The selloff wasn’t pure panic. It landed on a sector already showing cracks.
The Job Question
Markets weren’t the only thing trembling. The same week, Anthropic CEO Dario Amodei’s earlier statement resurfaced widely: that AI could displace half of all entry-level white-collar jobs in the next one to five years.
Salesforce CEO Marc Benioff piled on by announcing a halt to hiring for engineers, lawyers, and customer support roles, citing AI capabilities. And the numbers are already moving: in 2025, companies attributed 55,000 job cuts to AI — twelve times the number from two years earlier.
Piper Sandler downgraded Adobe, Freshworks, and Vertex, warning that “seat-compression and vibe coding narratives could set a ceiling on multiples.”
The Skeptics
Not everyone is panicking. Nvidia CEO Jensen Huang called the fears “illogical,” arguing that AI creates demand for new software rather than destroying it. Some analysts point out that the legal plugin still requires human oversight, that AI error rates remain significant in high-stakes work, and that enterprise adoption cycles take years, not days.
Bloomberg’s opinion desk called the doom-mongering “bizarre,” noting that every technology cycle produces a period of fear-driven repricing that eventually corrects.
There’s a historical case for optimism. ATMs were supposed to eliminate bank tellers. The number of tellers actually increased as banks opened more branches because ATMs made branches cheaper to operate. Spreadsheets didn’t eliminate accountants — they made accounting work more complex and valuable.
But those analogies have a shelf life. ATMs couldn’t draft legal briefs.
What This Means
The SaaSpocalypse isn’t really about one plugin. It’s about the market processing a structural shift that’s been building for months.
When AI could write code and generate images, investors treated it as a productivity tool that would help software companies ship faster. When it started automating the workflows those companies sold access to, the narrative flipped. AI went from growth driver to competitive threat.
This doesn’t mean every SaaS company is doomed. Companies with deep data moats, regulatory advantages, or workflows too specialized for general-purpose AI will likely survive. But the era of charging per-seat for software that an AI agent can replicate is entering its final chapters.
Stephen Yiu, CIO of Blue Whale Growth Fund, put it plainly: “This year is the defining year whether companies are AI winners or victims.”
For software companies, the answer increasingly depends on whether they’re building on top of AI or standing in its path.