Author: Cosmic Wave Naruto, Deep Tide TechFlow
In February 2026, the tech stock market is experiencing a systemic collapse referred to by some media as the "SaaSpocalypse."
Salesforce’s stock price has dropped nearly 40% from its peak in 2025; ServiceNow's stock plummeted over 11% in a single day after its quarterly earnings report, all due to management mentioning in a conference call that "AI agents are complicating the visibility of seat growth"; Workday fell over 22%; the entire S&P 500 software and services index has lost nearly $1 trillion in market value within the first six weeks of 2026.
The market logic is straightforward: AI agents are already capable of replacing a large number of manual operations, and companies using AI have completed work that previously required 100 people, so naturally, they no longer need 100 software seats. The seat-based pricing SaaS business model is believed to have reached its historical endpoint.
As this panic trading sweeps across the industry, Stephen Bersey, Head of U.S. Technology Research at HSBC, published a provocatively titled research report: "Software Will Eat AI."
His core argument can be summed up in one sentence: "The market's panic is a misjudgment."
The Contrarian Report
"Concerns that AI will replace enterprise software are misguided."
He writes at the beginning of the report. In his view, AI will not eliminate software but will be absorbed by it, becoming a capability layer embedded within enterprise software platforms. Software is not AI's opponent; software is the vehicle through which AI reaches the real world.
This logic flips the entire narrative framework of the current market. The market's fear is that "AI will replace software," while Bersey's assessment is that "software will tame AI."
He cites a historical analogy from the internet era: when the internet exploded, the initial value accumulation was concentrated in physical infrastructure—servers, fiber optic cables, data centers. A significant amount of capital flooded into hardware infrastructure, while those struggling early internet companies ultimately became the ones that gained long-term value. Software is the endpoint of internet value.
The evolution of AI, Bersey believes, is replaying the same script. The years 2024 and 2025 are the construction period for infrastructure, computational power, model integration—all of this is paving the way for an explosion at the software layer. And 2026 is the year when the engine really ignites.
"Software will be the primary mechanism for AI's diffusion across the world's largest enterprises. We believe 2026 will be the year when software monetization starts."
Why Can't Foundation Models Replace Enterprise Software?
The most substantial argument of the report systematically deconstructs the notion that "AI will directly disrupt software."
The critics' viewpoint appears persuasive: large language models can already write code, vibe coding (generating usable software directly from natural language descriptions) is on the rise, and AI model companies are making more applications at the layer above, so why do enterprises still need costly traditional software systems like Oracle, SAP, and Salesforce?
Bersey's answer unfolds across three dimensions.
First, foundation models have "inherent flaws."
The report clearly states that foundation models "have intrinsic defects" and cannot competently perform "complete replacements" of large enterprises' core platforms. They excel in narrow scenarios such as image generation, small application development, and text processing, but for high-fidelity, enterprise-grade core platforms, this is "not realistic."
The fundamental reason lies in the limitations of training data. LLMs are trained on publicly available internet data, whereas the proprietary architectural knowledge, business logic, and operational specifications accumulated by enterprise software systems over decades—these core intellectual properties are not available online, so AI cannot learn or replicate them. Oracle and SAP's moat is not something that can be caught up to by writing code; it is built over time and business scenarios.
Second, the capabilities of vibe coding are severely overestimated.
The report directly points to a critical weakness of vibe coding: it places the responsibility and burden of design entirely on the developer. If you tell AI "I want a system that can handle global supply chains," AI can generate code, but determining "how to define the architecture of this system, how to handle exceptions, and how to ensure it doesn't crash under extreme pressure" still requires human judgment.
More crucially, Bersey points out that those leading AI model companies "have almost no experience creating enterprise-grade software." They are starting from scratch in an incredibly complex environment. In contrast, enterprise software has evolved over decades to achieve "almost zero errors, high throughput, and high reliability"—a benchmark that AI newcomers cannot reach in the short term.
Third, the switching costs for enterprises are a real, high barrier.
Even assuming AI can write code of equal quality, the cost for enterprises to replace core systems remains extremely high: risks of revenue interruption, productivity losses, system compatibility issues across IT environments, and the trust built over time in vendor brands and service capabilities... All these are real switching costs that will not disappear because AI can write code.
Enterprise-grade software demands years of validated 99.999% uptime and error-free operation across various complex IT environments. This trust is earned over time, not produced by piling up code.
Who Will Be the Real Beneficiaries of AI Monetization?
If the first half of the report is defensive, the second half is an offensive strategy.
Bersey's core judgment is: the largest share of the AI value chain will ultimately flow to the software layer, not the hardware and chip layer.
"We believe AI is the primary source of value creation within the software stack, and the largest share of long-term value will belong to software, not hardware."
He also notes that hardware scarcity, GPU shortages, power limitations, and data center bottlenecks will continue to persist in the coming years. This scarcity reinforces the strategic position of software platforms: only software platforms can translate AI capabilities into scalable and repeatable business value.
The specific monetization vehicle pointed to in the report is AI agents.
Bersey predicts that by 2026, there will be large-scale deployment of task-oriented, workflow-embedded AI agents in Fortune 2000 companies and small to medium-sized enterprises. However, his qualitative definition of agents starkly contrasts with mainstream narratives in the market; he does not view agents as disruptors replacing software but rather believes agents must operate within the parameters and permissions defined by software. It is this "bounded agent" that can meet enterprises' needs for managing AI risks.
In other words, enterprises do not need an omnipotent, free-roaming AI; what they need is an AI that can be governed, audited, and operates within a compliance framework. This is only achievable by AI agents deeply embedded in enterprise software systems.
"Software is the key pathway for enterprises to controllably use AI." This is the core judgment of the entire report.
Additionally, the report predicts that inference demand will gradually surpass training demand, becoming the primary driving force behind computational consumption growth, which means that with the proliferation of agents, computational consumption will not shrink but will continue to grow, further supporting the entire software and infrastructure ecosystem.
Opportunity or Trap?
At the time of the report's release, the overall valuation of the software sector had already fallen to historical lows. Bersey's judgment is: the combination of undervaluation and the forthcoming monetization year is an entry opportunity, not a signal to exit.
"Software valuations are at historic lows, even as the industry is on the eve of massive expansion."
In terms of specific targets, HSBC's logic is clear: those companies that have established deep data moats, can incorporate AI agent capabilities, and do not rely solely on a headcount billing model will be the biggest beneficiaries of this wave of AI monetization. The buy rating list includes Oracle, Microsoft, Salesforce, ServiceNow, Palantir, CrowdStrike, Alphabet, among others, almost covering all core players in enterprise software.
It is worth noting that HSBC also downgraded IBM and Asana, placing Palo Alto Networks on "Reduce," indicating that not all software companies can safely navigate this storm; the key lies in whether they can become the infrastructure for AI agents' implementation and not merely a human interface that can be bypassed by agents.
Bersey's report is logically rigorous, timely, and its contrarian stance itself has a compelling communicative effect.
However, there is one question the report does not answer directly: if AI agents can indeed operate efficiently within the framework of enterprise software, will the demand for software "seats" quietly diminish? The value of software as a carrier for AI may hold, but whether the "per seat billing" business model can sustain the current valuations remains an open question.
Will software consume AI, or will AI consume software? This debate will be evidenced in every earnings report throughout 2026.
免责声明：本文章仅代表作者个人观点，不代表本平台的立场和观点。本文章仅供信息分享，不构成对任何人的任何投资建议。用户与作者之间的任何争议，与本平台无关。如网页中刊载的文章或图片涉及侵权，请提供相关的权利证明和身份证明发送邮件到support@aicoin.com，本平台相关工作人员将会进行核查。
