Only weeks after its latest workplace AI tools sent tremors through software stocks, Anthropic is pressing further into the enterprise world, expanding its ambitions inside offices and corporate workflows.
The company announced on Tuesday that it is upgrading its Claude AI assistant to perform more effectively across specific professional roles, including design, human resources, wealth management, and operations. Claude will also now be able to work directly inside commonly used applications such as Microsoft Excel and PowerPoint, rather than operating solely as a standalone chatbot. The updates were unveiled during a virtual event hosted by the company.
Anthropic first introduced Claude Cowork in January, positioning it as an evolution of its popular coding-focused AI into a broader workplace assistant. Since then, the company has rolled out a steady stream of upgrades. It began by launching plugins designed for tasks like legal research and financial analysis, followed by an improved underlying AI model and a specialized cybersecurity tool. With the latest update, Anthropic is now expanding Claude’s reach even deeper into everyday office work, with tools tailored to specific industries and job functions.
That rapid pace of development has unsettled investors. Wall Street has grown increasingly anxious that AI tools like Claude could quickly make existing enterprise software products obsolete — and potentially trigger large-scale job losses. Anthropic rejects that narrative outright.
According to Scott White, Anthropic’s head of enterprise product, the company is not attempting to replace established enterprise software platforms that businesses rely on to manage sensitive and proprietary data. Instead, Anthropic views its role as complementary.
“We see ourselves as a platform, not a product trying to own every workflow,” White said, emphasizing that Claude is designed to integrate with — rather than supplant — the software companies already trust.
Still, Anthropic’s urgency is not accidental. The AI sector is crowded, competitive, and moving at breakneck speed. Chief among its rivals is OpenAI, which has been aggressively expanding its own enterprise-focused offerings.
How Claude’s new workplace tools function
Rather than requiring users to switch to a separate chat interface, Claude can now operate directly within enterprise applications. It can pull in relevant context and data from documents, spreadsheets, and emails without forcing users to leave the task they are working on.
For example, Claude could analyze spreadsheet data inside Excel and automatically transform that information into a structured PowerPoint presentation — mirroring how a human employee might approach the task.
This approach eliminates the need to constantly copy and paste information between programs. White argues that this makes Claude feel less like a chatbot and more like “a real, fully capable virtual collaborator.”
Anthropic is also rolling out new plugins designed for specific professional fields. These tools allow Claude to model investment scenarios for private equity teams, draft job descriptions and offer letters for HR departments, create creative briefs for designers, and summarize vendor proposals for operations teams.
To develop its financial services tools, Anthropic worked with companies such as FactSet, S&P, and LSEG, while Apollo collaborated on private equity-focused capabilities.
Organizations will also be able to customize these plugins to integrate with their internal software stacks, including platforms like Google Drive, Gmail, Google Calendar, DocuSign, and others.
Software stocks feel the shock
The announcement follows a quieter rollout earlier this month of several industry-specific Claude Cowork plugins — a move that rattled software markets. In early February, a major software industry ETF fell nearly 6% in a single session, its worst daily performance since April.
Several prominent companies saw sharp declines. Thomson Reuters recorded its largest single-day drop on record, falling almost 16%. LegalZoom plunged nearly 20%. FactSet slid more than 10%, while European data analytics firm RELX lost about 14% in one day.
Even IBM was affected. Its shares dropped on Monday after Anthropic published a blog post outlining how AI could modernize COBOL — a decades-old programming language still used widely in business data processing. IBM currently sells tools designed to compile and upgrade COBOL code.
Cybersecurity stocks also dipped after Anthropic introduced a new Claude Code feature that scans entire codebases for vulnerabilities and recommends targeted software patches for human review.
A crowded race toward enterprise AI
Anthropic is not alone in its ambitions. OpenAI last month launched Frontier, a platform aimed at helping companies build, deploy, and manage AI agents capable of performing real business tasks. Earlier this week, OpenAI announced new multi-year partnerships with four major consulting firms, embedding OpenAI engineers directly within those organizations to help roll out its enterprise products.
The strategy appears designed to accelerate adoption by leveraging consulting firms’ access to large corporate clients.
Despite the hype and heavy investment, skepticism remains. Jacob Bourne, a technology analyst at eMarketer, previously told CNN that security and data protection concerns could slow widespread adoption of AI tools inside companies.
“Panic over this is probably misplaced,” Bourne said. “But it does mean that legacy enterprise software providers are going to have to keep evolving.”
As Anthropic continues to release updates at breakneck speed, one thing is becoming increasingly clear: whether businesses are ready or not, the AI race inside the workplace is accelerating — and it’s making investors, competitors, and entire industries uneasy.
