AI Business

Why Public Markets Misread the AI Threat to SaaS

Herd mentality and emotional trading wiped hundreds of billions from software valuations in early 2026, revealing a fundamental disconnect between market sentiment and actual business value in the AI era.

·4 min read
Public Markets Are Frequently Wrong. The SaaSpocalypse Shows Us Why
Public Markets Are Frequently Wrong. The SaaSpocalypse Shows Us Why

Collective panic and cascading sell-offs remain hallmarks of public market behavior, where investor sentiment can obliterate company valuations in moments. What often passes for price discovery becomes instead a reflection of crowd psychology rather than rational assessment of worth.

This dynamic surfaced vividly around artificial intelligence's potential impact on software firms and predictions of Software as a Service collapse. The so-called SaaSpocalypse generated considerable alarm, yet most market participants struggled to differentiate between software companies positioned to capitalize on AI and those genuinely threatened by it.

The episode did not demonstrate that software itself was doomed. Instead, it exposed how poorly equipped public markets are to assess value in rapidly shifting technological landscapes.

This pattern of misjudgment corrodes confidence in public equity markets. For companies weighing their future, public ownership may lose its appeal, and founders contemplating exits may no longer view an IPO as the ultimate achievement.

How the markets got it wrong

Richard Anton is the co-founder and general partner of Oxx
Richard Anton, co-founder and general partner of Oxx.

Public markets have struggled to process technological advances in the AI era and evaluate their consequences for established software businesses. Responses have ranged from insufficient—ignoring significant developments—to excessive, triggering unfounded alarm.

When OpenAI released ChatGPT publicly in November 2022, Alphabet faced market punishment despite ultimately establishing itself as a generative AI leader. Meanwhile, software markets moved sluggishly in absorbing the implications of generative and agentic AI capabilities through the end of 2025.

Enterprise software valuations continued climbing during this period until January 2026, when conditions deteriorated sharply. The introduction of Claude Cowork sparked widespread selling, intensified when supplementary integrations rolled out later that month targeting specific enterprise workflows in human resources and legal departments.

Valuations subsequently collapsed as traders applied blanket pessimism across the sector. Within forty-eight hours, approximately $285 billion in market value evaporated from software, data analytics and financial technology stocks. By April, the BVP Nasdaq Emerging Cloud Index had fallen 31% from the year's start.

Having invested in software across multiple decades, this pattern recurs regularly. Artificial intelligence does not signal software's demise. Rather, it has reshaped how software gets developed, where user expansion originates—primarily through agents rather than human users—and what creates lasting competitive moats. This transformation was always destined to reward certain companies, particularly those woven into customers' essential operations, while damaging others.

The recovery has been uneven and markets still get it wrong

Declarations of SaaS extinction followed the predictable arc of technology adoption cycles: initial euphoria, subsequent pullback, then renewed productivity gains. Markets initially overestimated the damage AI would inflict across the entire software landscape.

Correction is now underway as markets begin distinguishing which software companies gain from AI and which face headwinds. By late October, the cloud software index had rebounded 26% from January and surged 83% from its April low point.

At Oxx, our firm never accepted the SaaSpocalypse narrative, and markets have since recovered from the turmoil. However, the rebound remains uneven, with certain sectors like cybersecurity and data analytics performing strongly while others, including workflow management, have retreated.

Cybersecurity illustrates this divergence. CrowdStrike's stock fell 27% early in 2026 but climbed 114% between the year's beginning and late October.

The problem with public markets is part of a bigger trend

Reflexive market responses to AI developments represent more than temporary sentiment swings. They reflect a structural shift originating in the late 1990s and accelerating after 2002, when the Sarbanes-Oxley Act took effect to strengthen investor safeguards following the dot-com collapse.

An unintended consequence emerged: ownership migration from public to private markets. The U.S. public company count has halved while private markets have expanded substantially. This arena houses much of the sophisticated capital and a more measured, long-term perspective on software valuations.

The SaaSpocalypse panic demonstrated that public markets fail to accurately reflect the underlying value of numerous software businesses. This gap may be generating systematic distortions in private transaction pricing—multiple deals underway early in the year collapsed after public markets sent false signals. This misalignment requires attention. Currently, few acknowledge the dysfunction in public market price discovery, and recognizing the problem itself represents the essential first step toward resolution.

Richard Anton is co-founder and general partner at Oxx, a software investment firm backing European founders pursuing industry transformation through AI as they prepare to scale.