Software stocks have had a bruising few months. High-profile names such as Salesforce (-27%) Workday (-37%), and DocuSign (-33%) have all come under pressure as investors reassess what artificial intelligence might mean for traditional “seat-based” software models. Even mega-cap leaders like Microsoft (-17%) have not been immune to the volatility.
The phrase quietly circulating on Wall Street is “SaaSpocalypse” (SaaS – software as a service + apocalypse) the idea that AI agents could automate large portions of white-collar work, reducing corporate headcount and, in turn, the number of software licenses companies need to buy. If AI can draft emails, write code, prepare reports and analyse data, why pay for as many human seats?
That fear intensified after a widely circulated “thought exercise” memo by Citrini Research was released last week, which outlined a rapid AI adoption scenario leading to significant white-collar displacement and as a result plummeting stock and house prices due to mass unemployment. Markets reacted first and debated later, erasing hundreds of billions in market value across software companies in a matter of days.
The concern is as follows. Many software firms have built their growth models on expanding enterprise headcount. More employees meant more subscriptions. If AI compresses labor demand, revenue growth could stall.
To explain with an example. The contrast between Microsoft and Salesforce illustrates the market’s dilemma. Microsoft has embedded AI across its ecosystem from Office to Azure cloud services. Its opportunity lies not just in selling productivity seats (Microsoft Word, Excel licenses etc, but in monetising AI usage and computing consumption through Azure Cloud. Therefore, Microsoft benefits from growing AI demand volumes that will offset any lower margins expected on the productivity or selling seats side of their business.
Salesforce, by contrast, historically scaled alongside enterprise hiring. Its new AI push signals a shift from software that helps humans work, to software that can replace certain human tasks. That transition is innovative but it complicates Salesforce revenue generation going forward. In the short term, Salesforce customers, because of AI, will be able to do more with less workers.
The broader technology market narrative reinforces this divide. Infrastructure providers such as Nvidia and hyperscale cloud operators are perceived as early winners, supplying the computational backbone of the AI buildout. Application-layer software firms face more scrutiny, as investors debate whether AI agents commoditise traditional software interfaces.
Yet history urges caution before declaring extinction events.
The Jevons Paradox
Technological revolutions have repeatedly triggered similar fears. The introduction of electronic spreadsheets in the late 1970s was expected to eliminate finance roles instead, it expanded financial analysis and corporate forecasting. Automated teller machines (ATMs) were predicted to wipe out bank teller jobs but in practice led to lower operating costs and allowed banks to open more branches and expand services. The internet was supposed to destroy retail and media outright but instead, reshaped and expanded them.
This pattern reflects what economists call Jevons Paradox: when technological efficiency improves, costs fall and, counterintuitively, total consumption or production often rises rather than falls. If AI dramatically reduces the cost of producing analysis, writing code, or generating content, the volume of those activities may surge. Intelligence, in effect, becomes cheaper and more widely used.
None of this dismisses disruption risk. AI will alter labor markets. Some roles will shrink and others will emerge. Revenue models tied purely to headcount growth may need to evolve. Transition periods can be volatile, particularly if companies hesitate to invest while rules of the game are being rewritten.
Current valuations help to paint the fear picture investors are currently grappling with. Microsoft currently has a price to earnings ratio of 22 while Walmart trades at 43. Walmart based on earnings is twice as expensive as Microsoft. Which company is likely to grow quicker over the next 10 years?
Cayman investors need to assess AI disruption risk in their portfolio and which companies are either immune from this technology advancement or are actively investing to integrate it.
For Cayman businesses, particularly in financial and professional services, this period is less about existential risk and more about experimentation: identifying where AI deployment can lift productivity, reduce administrative inefficiency and ultimately drive larger economic output.
Disruption is rarely painless, but history shows economies adapt. Productivity shocks have tended to expand overall output, not permanently shrink it. If AI follows that pattern, today’s volatility may reflect uncertainty rather than collapse. As Amara’s Law reminds us, we tend to overestimate the impact of new technologies in the short run and underestimate them in the long run. AI may prove no different.
Robert Whelan, a chartered accountant, is the portfolio manager at NCB Capital Markets (Cayman) Ltd.
