The Software Sector's Rebound After an AI Scare
For much of the past year, the software sector found itself caught in a wave of pessimism. Investors broadly assumed that artificial intelligence would prove deeply disruptive to enterprise software incumbents, threatening the business models of companies that had long dominated the category. That fear pushed the sector lower, with broad-based software indices lagging significantly behind other parts of the market. More recently, however, the narrative has begun to shift. The leading software benchmark has climbed roughly 8 to 9 percent in a single month as investors refocus their attention on fundamentals heading into earnings season.
The setup is favorable. Over the past year, many software companies guided conservatively, establishing low expectations that now appear beatable. Just as importantly, AI itself is becoming a tool these companies use to lower their own cost of delivery. Software firms employ more engineers than almost any other kind of business, and the engineering profession is arguably the single occupation whose productivity has been most dramatically enhanced by AI. That means these companies can build and ship their products for considerably less than before, with obvious implications for the bottom line. The stock performance we're seeing reflects that improved earnings outlook, with names like Oracle, Palantir, IBM, Salesforce, and Microsoft all participating in the move.
The Rise of Third-Party Agents in the Enterprise
A deeper shift is also underway in how enterprise software is actually consumed. Until recently, corporate software packages were designed for human operators sitting at a screen. Over the past several weeks and months, that assumption has begun to break down. Enterprise software is increasingly being operated not by employees but by AI agents — and critically, these are often not the agents built by the software vendors themselves. They are third-party agents coming primarily from Anthropic and, to a lesser extent, OpenAI.
This dynamic reframes the entire landscape. A platform like ServiceNow, long organized around IT service management workflows for human users, must now adapt its offering so that external agents from frontier AI labs can operate it effectively. That is a nontrivial engineering and product challenge, but so far ServiceNow has navigated it well. If the company continues on that trajectory, the shift could actually become a tailwind rather than a threat. As long as the major AI labs don't absorb all the available attention and budget, well-designed packaged software becomes more valuable, not less, because the agents need something capable to operate on top of.
For ServiceNow specifically, the fundamentals support continued ownership into and through earnings. The company looks capable of growing in the mid to high teens for several years while simultaneously expanding margins and adapting to the AI era. The valuation has also become more reasonable. A year ago the stock was trading on revenue multiples — roughly 15 times sales — but today it trades at less than 20 times cash flow, a materially more attractive level given the overhang of AI-related concerns across the sector.
IBM's Diversified Exposure to the AI Buildout
IBM offers a different but generally constructive case. Its diversified portfolio spans hardware, software, and services, and several of those lines have positive exposure to the AI transition. Artificial intelligence cannot actually be deployed in the enterprise without someone implementing it, and it cannot run without infrastructure software underneath it. On both the hardware and software sides of its business, IBM is well positioned to benefit from that buildout.
The more complicated piece of IBM's story is the services business. If AI enables consulting and implementation work to be delivered at lower cost, the net impact on a legacy services franchise becomes harder to predict. Lower unit economics could translate into pricing pressure, or they could translate into volume expansion as more clients pursue projects that were previously uneconomic. The resolution of that trade-off is one of the central questions for IBM's trajectory.
Apple's Succession Signals a Hardware-First Future
The leadership transition at Apple tells its own story about where the company believes value will be created next. Tim Cook is exiting at a high, but that also means his successor, John Ternus, inherits a company at a peak, with all the difficulty that implies for sustaining momentum. The board's choice of Ternus — a hardware leader — is revealing. It suggests Apple believes its path forward has more to do with hardware innovation than with software, services, or direct participation in the frontier model race.
That logic is coherent. The way humans interact with computing is changing rapidly. The graphical user interface on desktops and the touch interface on handsets are giving way to voice-based interaction with AI, and AI systems are also gaining the ability to perceive the world through cameras and video. These new modalities call for new form factors. The next step for Apple might be something relatively incremental, such as a folding phone, following competitors that have already entered that category. Beyond that, however, the company's hardware ambitions likely extend into smart glasses, AI-native wearable pins, and a renewed push on the spatial computing platform that the Vision Pro introduced. All of those are fundamentally hardware bets, which is precisely why a hardware leader was chosen to run the company.
A Potentially Historic Year for Tech IPOs
If current timelines hold, this could become a banner year for public market debuts in technology. SpaceX is expected to come to market, potentially alongside Anthropic or OpenAI in roughly the same window. Each of these companies is likely to command a valuation of at least one trillion dollars at the point of initial trading — unprecedented in scale and in the sheer volume of capital required.
The economic significance of these businesses is difficult to overstate. Anthropic and OpenAI combined are already operating at an annual run rate above $50 billion, having started from essentially zero only two or three years ago. Taken together, that figure already exceeds the revenue of every software company in the world other than Microsoft. Within a year or two, the two labs are likely to be the second and third largest software companies by revenue. Their public listings will not simply be significant financing events; they will reshape public market indices, capital allocation patterns, and the competitive landscape for every other company building around AI.
The Larger Picture
The common thread across these stories is that AI is simultaneously a threat and a tailwind for the software ecosystem, and the winners are the companies that can host the agents, furnish the infrastructure, or pivot their hardware to new interaction modalities. The initial investor reflex — that AI would simply destroy incumbents — is giving way to a more textured view in which well-positioned software platforms become more valuable in an agent-driven world, diversified technology conglomerates benefit from implementation demand, and hardware leaders bet on a post-touchscreen future. The IPO pipeline on top of all of this suggests that the capital markets are preparing to absorb a once-in-a-generation wave of AI-native listings, with consequences that will extend well beyond the technology sector itself.