A broad and unusually swift selloff has swept through global technology stocks this week, erasing hundreds of billions of dollars in market value across equities, bonds and leveraged loans tied to the sector. What makes this episode distinct from earlier AI-related volatility is the trigger: not concerns of a speculative bubble, but mounting evidence that artificial intelligence may be reaching a stage where it starts to displace the very software and service models that powered the last two decades of technology growth.

Software companies were at the epicenter. An iShares ETF tracking global software names has shed close to a trillion dollars in value over the past week, underscoring the scale and speed of the repricing. The selling has since broadened into chipmakers, cloud infrastructure players, private equity-backed software firms, and even lenders exposed to the sector through technology loans that have now slipped into distressed trading territory.

The immediate spark came from a seemingly modest announcement by AI startup Anthropic, which unveiled a legal-focused tool capable of reviewing contracts and performing high-value documentation tasks. In isolation, the product may not appear transformational. But in the context of rapid AI advancements in coding, documentation, research and automation over the past year, markets interpreted the launch as another signal that AI is steadily encroaching on knowledge-intensive work once thought secure.

Analysts noted that if AI tools can reliably perform legal review today, similar applications could quickly emerge for sales documentation, marketing workflows, financial analysis and customer support core functions serviced by a vast global software ecosystem.

This fear has coincided with discomforting signals from companies long viewed as the primary beneficiaries of the AI boom. Alphabet Inc. indicated that its capital expenditure on AI infrastructure will be significantly higher than earlier estimates, raising questions about returns on investment. Arm Holdings Plc issued a revenue outlook below expectations, hinting that AI optimism is not yet translating into near-term commercial traction. Both developments weighed on sentiment further.

The selling quickly transcended US markets. Shares of London Stock Exchange Group Plc, Tata Consultancy Services Ltd, and Infosys Ltd declined during the week as investors reassessed the vulnerability of traditional IT and software service models to AI-driven automation.

Asian markets were not spared either. Weakness in Samsung Electronics Co. dragged South Korea’s benchmark lower, while concerns around Arm’s outlook affected shares of SoftBank Group Corp. in Tokyo. The pattern indicates a cross-market repricing of companies linked, directly or indirectly, to software, semiconductors and AI infrastructure.

Importantly, this selloff is unfolding despite the absence of visible operational damage so far. Major enterprise software providers have not reported customer attrition due to AI tools. Companies such as ServiceNow and Salesforce have not flagged earnings misses attributable to AI substitution. Yet, markets appear to be forward-pricing a scenario where AI compresses billing models, reduces headcount requirements, and shortens project cycles all of which could pressure revenue growth for IT service providers and software vendors alike.

Evidence of slow monetization of AI tools is also adding to investor caution.

Microsoft Corp. recently disclosed that its Copilot product has about 15 million paying users, a small fraction of its overall user base. This raises questions about how quickly AI features can translate into scalable revenue streams versus becoming productivity tools that reduce billable work.

For Indian IT majors, the development is particularly sensitive. Companies such as TCS and Infosys derive a substantial portion of revenue from application development, maintenance, testing, documentation, consulting and support services precisely the categories where AI tools are showing early capability. Even if clients do not immediately reduce spending, future deal structures, pricing benchmarks and staffing models may change.

The fear is not of an abrupt collapse in demand, but of gradual margin compression and slower growth as AI-enhanced clients demand more output with fewer billed hours.

From a credit market perspective, the stress is already visible. Over $17 billion worth of US tech company loans have slipped into distressed levels in recent weeks, indicating that leveraged exposure to software companies is being repriced sharply. This adds a financial stability angle to what began as an equity market story.

Market strategists argue that the selloff is self-reinforcing. As valuations fall sharply, momentum traders and institutional funds cut exposure, which amplifies the decline and spreads it across related sectors, regardless of individual fundamentals.