The software industry, long considered one of the most durable profit engines in the history of capitalism, is in the middle of a reckoning that few on Wall Street saw coming. The iShares Expanded Tech-Software Sector ETF (IGV) has plunged more than 22% from its recent highs, officially entering bear market territory. The damage in dollar terms is staggering: nearly $1 trillion in market capitalization has evaporated from software stocks in a little over a month.
The sell-off is not the product of a broad economic downturn, a sudden interest rate shock, or a regulatory assault. It is the result of something far more existential: the growing realization that artificial intelligence, the technology that was supposed to supercharge software companies, may instead be eating them alive.
The Numbers Are Brutal
Of the 110 stocks tracked in the iShares software ETF, 100 have posted negative returns so far in 2026. More than 20 have declined by over 30%. The casualties include some of the most celebrated names in enterprise technology. ServiceNow, the workflow automation giant that was once considered a best-in-class SaaS compounder, has tumbled 28% year to date. Salesforce, the company that popularized cloud-based customer relationship management, has fallen nearly 26%. Intapp, a professional services software maker, has cratered 49.7%.
The breadth of the decline is what makes it so unnerving. This is not a story about one or two companies missing earnings expectations. It is a sector-wide repricing that reflects a fundamental reassessment of whether the traditional software-as-a-service business model can survive in a world where AI agents can perform many of the same tasks that enterprise software currently automates.
"What we are witnessing is not a typical sector rotation. It is the market pricing in the possibility that AI will compress the revenue potential of an entire category of technology companies. The infinite-margin SaaS model is being challenged by the very technology it was supposed to benefit from."
Dan Ives, Managing Director, Wedbush Securities
Why AI Is a Threat, Not Just an Opportunity
For years, the conventional wisdom on Wall Street was that artificial intelligence would be an unambiguous positive for software companies. The logic was straightforward: AI features would increase the value of existing software products, justify higher prices, and reduce customer churn. Companies like Microsoft, Salesforce, and ServiceNow raced to integrate AI capabilities into their platforms, often charging premium prices for the new functionality.
The problem, it turns out, is more nuanced. AI tools are advancing so rapidly that they are not merely enhancing existing software products. They are, in some cases, replacing them entirely. When an AI agent can draft a customer email, schedule a follow-up, update a CRM record, and analyze the sales pipeline in a single automated workflow, the value proposition of a standalone CRM platform begins to erode.
The trigger for the latest wave of selling was the unveiling of increasingly capable AI coding and task-completion tools that can perform complex professional workflows, the very tasks that many enterprise software companies charge tens of thousands of dollars per year to automate. The implication was immediate and brutal: if AI can replicate the core functionality of expensive SaaS tools at a fraction of the cost, then the pricing power that underpinned the software industry's extraordinary margins is at risk.
The Margin Problem Nobody Anticipated
Even software companies that have successfully integrated AI into their products are discovering an uncomfortable economic reality. Approximately 70% of software providers now acknowledge that the cost of delivering AI features is eating into their profitability. The culprit is GPU compute, which is expensive, supply-constrained, and growing more so as demand from large language model training competes with inference workloads.
The SaaS business model was built on the premise of near-zero marginal cost. Once a software product was developed, each additional customer added almost pure profit to the bottom line. Gross margins of 75% to 85% were standard, and the best companies pushed above 90%. AI changes that equation fundamentally. Every AI-powered query, every automated workflow, every intelligent recommendation requires compute resources that cost real money. The era of infinite margins may be over.
The Contagion Is Spreading
The software sell-off has now begun to infect adjacent sectors of the financial markets. Private credit firms with heavy exposure to software company lending have seen their stocks decline sharply. Blue Owl Capital fell as much as 13% in a single session, while vehicles managed by Ares, KKR, Blackstone, TPG, and Apollo Global Management declined between 8% and 10%. The software sector represents approximately 22% of total debt exposure in assessed private credit portfolios, or roughly $224 billion.
Ratings agency KBRA has warned that in an aggressive disruption scenario, default rates in U.S. private credit could climb to 13%, significantly higher than the stress projections for leveraged loans and high-yield bonds. The concern is not hypothetical: if software companies lose pricing power and see revenue decline, their ability to service the leveraged buyout debt that private equity firms loaded onto their balance sheets becomes increasingly precarious.
Winners and Losers in the AI Reshuffling
Not every software company is equally vulnerable. Companies with deeply embedded workflows, high switching costs, and proprietary data advantages are better positioned to weather the disruption. Microsoft, which owns both the dominant enterprise software suite and a leading AI platform through its partnership with OpenAI, has been relatively insulated. So have companies like Palantir, whose government and defense contracts create a moat that AI startups cannot easily breach.
The most vulnerable companies are those selling horizontal software tools, products that automate general business functions like scheduling, project management, and basic data analysis, where AI alternatives are most readily available. Vertical software companies that serve specialized industries with domain-specific workflows and regulatory requirements may fare better, though they are not immune.
For investors, the software bear market presents both danger and opportunity. Valuations have compressed dramatically, with many formerly high-flying SaaS stocks now trading at single-digit revenue multiples for the first time in years. Some of these companies will adapt, integrate AI into their products in ways that add genuine value, and emerge stronger. Others will see their business models permanently impaired.
What Comes Next
The key question for the rest of 2026 is whether the software sell-off is an overreaction driven by fear and narrative, or a justified repricing of structural risk. History suggests the answer is probably somewhere in between. Markets tend to overshoot in both directions, and the panic-driven selling of the past month has likely pushed some genuinely strong businesses to prices that do not reflect their long-term earnings power.
But the underlying shift is real. Artificial intelligence is not just a feature to be bolted onto existing products. It is a paradigm change that will redefine what software is, how it is priced, and who captures the value it creates. The $1 trillion evaporation in software stock market value is not the end of that story. It is the opening chapter.
