Anthropic Shakes Up Three Industries in Three Weeks, Leaving Wall Street Unprepared

Anthropic’s rapid‑fire product cadence has forced investors to reassess the competitive landscape of three traditionally insulated enterprise markets. On Jan. 31 the company unveiled Claude Cowork, an AI‑driven assistant that can ingest calendar data, parse emails, generate project plans and field customer‑service tickets without the need for a separate SaaS stack. Within 48 hours the iShares Expanded Tech‑Software Sector ETF slumped 7 percent, while ServiceNow, Salesforce and LegalZoom fell 7.6 percent, 7 percent and 20 percent respectively, according to the Moth report dated Mar 8. The market reaction signaled that investors now view a single, general‑purpose AI layer as a credible substitute for a suite of vertical SaaS applications that have long commanded premium pricing and sticky contracts.

Just three weeks later Anthropic introduced Claude Code Security, an autonomous agent that continuously scans an organization’s codebase, identifies exploitable vulnerabilities, and applies patches without human oversight. The product’s “inside‑the‑code” approach contrasts sharply with conventional security solutions that rely on external monitoring and rule‑based alerts. By Feb. 23, CrowdStrike’s shares had plunged 11.3 percent—the worst single‑day drop in its history—while Cloudflare and Zscaler each lost between 8 percent and 10 percent, and the Global X Cybersecurity ETF fell nearly 9 percent in a single session, as noted in the same Moth analysis. Palo Alto Networks CEO Nikesh Arora attempted to defend the human‑judgment element of traditional security during earnings on Feb. 18, yet his stock still fell 7 percent, underscoring that the market values the new operational paradigm over legacy arguments about threat‑intelligence expertise.

The third wave arrived on the same Monday, when IBM recorded its steepest trading loss since October 2000, with shares down 13.2 percent and a cumulative 27 percent decline for February—its worst month‑to‑date since at least 1968, per the Moth report. Anthropic’s Claude Code now maps dependencies across massive COBOL codebases, automatically documents workflows and flags legacy‑risk exposures that previously required weeks of manual analysis. COBOL still underpins 95 percent of ATM transactions and 80 percent of in‑person banking operations, and 45 of the 50 largest U.S. banks run critical workloads on the language. The consulting market built around COBOL modernization is estimated at over $30 billion annually, with firms such as Accenture and Cognizant charging $200‑$400 per hour for scarce senior engineers. Anthropic’s automation of this “knowledge‑capture” function effectively eliminates the scarcity premium that has sustained a lucrative consulting niche for decades.

Collectively, these three launches have not merely dented individual stocks; they have compressed an entire sector. The iShares Expanded Tech‑Software Sector ETF, which tracks a broad swath of software companies, is now down 23 percent year‑to‑date, reflecting a category‑wide repricing rather than isolated company failures, as the Moth analysis emphasizes. Analysts cited in the report argue that the sell‑off may be overdone, yet the ETF’s trajectory suggests that investors are pricing in a structural shift: AI agents that can replace horizontal SaaS platforms, internal security monitoring and legacy‑code consulting in a single, continuously learning system. The market’s response has been swift and decisive, indicating that Wall Street is still scrambling to quantify the long‑term revenue impact of Anthropic’s platform‑level capabilities.

The broader implication is a redefinition of “software” as a commodity that can be abstracted into a generative AI service layer. While Anthropic has not directly “crashed” any single company, it has destabilized the business models of entire categories in a matter of weeks. The firm’s aggressive rollout cadence, combined with the immediate market reaction documented across multiple ETFs and individual equities, suggests that future AI launches will be evaluated not just on incremental functionality but on their potential to supplant entire revenue streams. Investors and incumbents alike must now confront the prospect that the next wave of AI could render whole consulting practices and security product suites obsolete, a reality that Wall Street appears only just beginning to price in.