JP Morgan joins the ranks of the software stocks to support US stocks: Concerns about AI impact are exaggerated, and a rebound is expected after a historic decline.

JPMorgan strategist says that as the market has priced in unrealistic expectations for the short-term disruption of the software industry by artificial intelligence (AI), software stocks are expected to rebound from their historically low levels. Led by Dubravko Lakos-Bujas, the strategist team pointed out that due to the current "extreme price volatility", investors should increase exposure to high-quality software companies that are more resilient to AI, at least in the short term. The team wrote in a report, "Given that market positions have been fully cleared, prospects for the AI-disruption of the software industry are overly pessimistic, and fundamentals remain robust, we believe that the balance of risk-return is increasingly tilted towards a rebound". Due to concerns that new AI tools may impact traditional Software as a Service (SaaS) business models, U.S. software stocks have been under pressure recently. The sell-off did not distinguish between companies that have partnerships with AI companies or proprietary data assets, almost treating all related software companies equally. The software sector has fallen to its lowest level since the market turmoil of last April. Microsoft Corporation (MSFT.US) and CrowdStrike (CRWD.US) are among the representative companies with AI resilience mentioned by JPMorgan strategists, which are expected to benefit from improving workflow efficiency through AI. The team noted that the high conversion costs and multi-year contracts of enterprise software provide a buffer against short-term shocks. The JPMorgan team also pointed out that the uncertainty remains as to whether traditional software companies will be replaced by AI in the long term, but the current market sentiment towards AI disruption is overly pessimistic at this stage. They added that the overall performance of the software industry's fourth-quarter earnings reports is positive, with analysts forecasting a profit growth rate of 16.8% for the industry by 2026. This bullish view is also echoed by Morgan Stanley's strategist team led by Michael Wilson. The strategist team stated this week that there is still further upside potential for U.S. technology stocks, and the decline in software stocks has opened an "attractive entry window". Wilson wrote in a report, "Such volatility in major investment cycles is not uncommon. Despite this, the fundamental advantages of AI-empowered companies still exist, and we believe that the trading value of AI adopters is still undervalued". Last week, analysts led by Wedbush's Dan Ives, known as the "tech bull", also stated that while AI may indeed pose a certain pressure on traditional software business models in the short term, the market's reaction to this risk is clearly excessive. The current sell-off of software stocks implies an extreme assumption that the industry will be massively disrupted by AI, which is not feasible in reality. Ives pointed out that enterprise clients are far more cautious in the AI migration issue than the market imagines. Many companies are not willing to expose their core data to immature new platforms to chase after AI dividends, nor are they willing to easily give up the software infrastructure built over the past decades at a cost of billions of dollars. He said, "AI is a headwind in the short term, no doubt about it, but the current market pricing suggests that the software industry is about to face a 'doomsday', a judgment that in our view is completely detached from reality". Wedbush emphasized that the current large enterprise software ecosystem has accumulated trillions of data points, and emerging AI companies such as OpenAI and Anthropic are unlikely to completely take over these complex systems in the short term, whether in terms of data capacity or enterprise-level security. This means that AI is more likely to be integrated into existing software platforms in the form of "embedded tools" rather than completely replacing them. Wedbush also pointed out that Microsoft Corporation, Palantir Technologies (PLTR.US), CrowdStrike, Snowflake (SNOW.US), and Salesforce, Inc. (CRM.US) are the top five software stocks worth holding in the current "software cold winter". NVIDIA Corporation (NVDA.US) CEO Jensen Huang also refuted concerns that AI will replace software and related tools last week, calling such ideas "illogical". Huang stated during a speech at an artificial intelligence conference hosted by Cisco Systems, Inc. in San Francisco, that the idea that AI will diminish the importance of software companies is misleading. He believed that AI will continue to rely on existing software rather than reinventing basic tools from scratch. Huang said, "There is a view that the tools of the software industry are declining and will be replaced by AI...which is the most illogical thing in the world, time will prove everything." "If you are human or Siasun Robot & Automation, whether it's human-controlled or general-purpose Siasun Robot & Automation, would you use tools or reinvent them? The answer is clear, use tools...that's why the latest breakthroughs in AI are about the use of tools, as tools are designed to clearly perform their functions." AI reshapes the value chain and future outlook From the perspective of software engineering reality and SaaS industry structure, the narrative of "AI replacing the entire enterprise software stack" is an easily extrapolated story by the market. The "value density" of enterprise software lies not only in interfaces and functions, but also in proprietary data, permissions/auditing chains, compliance and responsibility boundaries, system integration, SLAs and availability, change management and organizational processes; which determine that even with the strongest large language models (LLMs), they often need high-quality proprietary corpora + structured knowledge bases + controllable tool calls + trackable outputs to operate in production environments. From the underlying technical logic of AI tools and SaaS software fields, panic selling of software stocks does not mean "software is no longer needed", but the value chain is being redistributed by AI: stronger general large models and agentic AI workflows make many "point-functional SaaS" systems face the risk of being internalized by the model layer/platform layer's functionalities or being bypassed by "conversation-oriented entry + automated execution" to avoid UI and seat-based pricing and renewal logic; thus, the market is more eager to split software stocks into "AI winners/AI losers". However, in turn, the "system record layer" (ERP/CRM/ITSM/databases/security/compliance) of enterprise SaaS software leaders often possess data sovereignty, governance, permissions, auditing, and migration cost barriers. The reality is more likely that AI turns these long-standing software giants into distribution channels for delivering AI capabilities rather than overnight replacements for the entire existing software infrastructure. According to the Morgan Stanley strategy team, at the short-term level, the sharp drop in software stocks has indeed triggered discussions about the "technical aspects approaching a stage bottom" and some funds have slightly increased their positions; however, more capital is still waiting for hard catalytic logic that can turn the narrative of AI applications into actual revenue curves such as software companies disclosing AI-related product revenue/penetration rates, enterprise clients announcing large-scale deployments, or renewal indicators (net retention, expansion rate) strengthening significantly after the introduction of large AI models or agentive AI intelligences; without this evidence, the rebound is more like "oversold recovery" rather than a new trend. The recent wave of software stock sell-offs is more like the market's extreme way of answering a new question: to what extent will the profit pool of SaaS software vendors be redistributed by "model factories + agents"? In the short term, the answer can only be verified by two "hard indicators": (1) the speed of actual deployment and payment diffusion at the enterprise end; (2) the elasticity of SaaS vendors' AI-related product revenue and renewal/net retention after introducing large models or agentive AI intelligences as some buyer representatives described in a part of the Thomson Reuters Corporation Breakingviews research report, they are waiting for "actual revenue growth data from AI-related products" or more enterprise deployment announcements as buying catalysts. Before this, the fluctuations in software stocks are likely to continue: on the one hand, technical aspects may see an "oversold rebound", while funds will continue to make structural switches preferring vertical software/data asset companies closely associated with AI training / reasoning systems and platforms; and those with weaker moats, higher homogeneity, and higher valuations in the application layer will continue to require higher risk premiums. Therefore, companies like Microsoft Corporation, MongoDB, Snowflake, Palantir, and SAP, which aggregate data assets and have strong fundamentals, may find it easier to rebound after the panic.