The artificial intelligence boom has created a stark dividing line across the software industry. On one side stand companies that are embedding AI into their products in ways that drive genuine revenue growth and customer retention. On the other are firms watching their legacy business models erode as AI-native competitors and shifting buyer expectations threaten to leave them behind. For investors, distinguishing between the two camps has become the central challenge of 2025.
Two prominent market strategists recently laid out frameworks for making that distinction, and their analysis carries significant implications for portfolio construction in the technology sector. According to reporting by MSN, the experts argue that the AI-driven bifurcation in software is no longer theoretical — it is showing up in earnings reports, guidance, and stock performance right now.
The Monetization Test: Who Is Actually Making Money From AI?
The core thesis is straightforward but powerful: the software companies that will outperform over the next several years are those that have moved beyond simply talking about AI integration to actually monetizing it. This means measurable increases in average revenue per user, new product lines generating incremental bookings, and — perhaps most telling — pricing power that stems from AI features customers cannot easily replicate or replace.
Microsoft remains the most frequently cited example of a company passing this test. Its Copilot suite, integrated across Office 365, Azure, and GitHub, has given the company multiple vectors for AI monetization. In its most recent quarterly earnings, Microsoft reported that Azure revenue growth re-accelerated, driven in significant part by AI workloads. CEO Satya Nadella told analysts that AI-related Azure revenue was on a $13 billion annual run rate, a figure that has roughly doubled in the past year. The stock has responded accordingly, outperforming the broader software sector in 2025.
ServiceNow, Palantir, and the New Tier of AI Beneficiaries
But the story extends well beyond Microsoft. ServiceNow has emerged as another clear beneficiary, with its Now Assist AI capabilities driving upsell opportunities within its existing enterprise customer base. The company reported in its April earnings call that AI was contributing to deal sizes expanding by 20% or more in some cases, a data point that analysts seized upon as evidence of real monetization rather than hype. CEO Bill McDermott has been vocal about positioning ServiceNow as an “AI platform company,” and the market has largely bought into that narrative.
Palantir Technologies represents a more polarizing case. The company’s stock has surged more than 60% year-to-date as of late June 2025, fueled by accelerating U.S. commercial revenue and high-profile government AI contracts. Bulls argue that Palantir’s Artificial Intelligence Platform (AIP) has created a new category of enterprise software that lets organizations deploy large language models on their own proprietary data. Skeptics counter that the valuation — now well above 100 times forward earnings — prices in years of flawless execution. Regardless of where one lands on valuation, Palantir is clearly in the “winner” column when it comes to AI-driven revenue momentum.
The Losers: Legacy Software Facing an AI Headwind
On the other side of the ledger, the experts highlighted several categories of software companies facing genuine existential pressure from AI. Traditional customer support and helpdesk software providers are among the most exposed. Companies like Zendesk, which went private in 2022, and publicly traded peers in the customer experience space are watching as AI-powered chatbots and automated resolution tools reduce the need for large numbers of human support agents — and, by extension, the number of software seats enterprises need to purchase.
The seat-based pricing model, long the backbone of SaaS economics, is under direct threat. If an AI agent can handle the work of five customer service representatives, an enterprise that previously needed 500 Zendesk licenses might now need 100. This dynamic is not limited to customer support. Companies selling tools for basic code generation, content creation, data entry, and routine analytics all face similar compression. As MSN reported, the experts emphasized that investors need to carefully examine whether a company’s AI features are additive to revenue or whether AI is actually cannibalizing the company’s core business.
The Valuation Puzzle: Paying Up for Growth That May Not Materialize
One of the trickiest aspects of the current environment is valuation. Many of the perceived AI winners are trading at substantial premiums to their historical averages. The BVP Nasdaq Emerging Cloud Index, which tracks high-growth cloud software companies, has seen its median forward revenue multiple expand significantly since the start of 2024, even as many constituent companies have yet to demonstrate meaningful AI revenue contribution.
This creates a dangerous setup for investors who buy into the AI narrative without scrutinizing the underlying numbers. Not every company that adds “AI” to its investor presentation is going to see a step-function improvement in growth. The experts cautioned that the market is likely to become more discriminating as 2025 progresses, rewarding companies with verifiable AI revenue metrics and punishing those that cannot deliver. Recent earnings seasons have already shown this pattern: stocks like CrowdStrike and Datadog, which have demonstrated clear AI-related demand drivers, have been rewarded, while companies with vague AI roadmaps have seen muted or negative post-earnings reactions.
The Infrastructure Layer: A Different Kind of Winner
Beyond application software, the infrastructure layer supporting AI workloads continues to attract enormous capital. Nvidia’s dominance in AI training and inference chips is well documented, but the software companies that enable AI infrastructure are also benefiting. MongoDB, for instance, has seen increased adoption of its database platform as companies build AI applications that require flexible, high-performance data storage. Snowflake, after a period of decelerating growth, has pointed to AI and machine learning workloads as a catalyst for re-acceleration in data consumption on its platform.
Cloudflare is another name gaining attention. The company’s Workers AI platform allows developers to run AI inference at the edge of its global network, and CEO Matthew Prince has described AI traffic as one of the fastest-growing workload categories on Cloudflare’s infrastructure. While still early in terms of revenue contribution, the strategic positioning has helped the stock recover from its 2022-2023 drawdown.
What Smart Money Is Doing Now
Institutional investors appear to be taking a barbell approach to AI software exposure. On one end, they are concentrating positions in proven AI monetizers — Microsoft, ServiceNow, CrowdStrike, and Palantir among them. On the other end, they are selectively betting on smaller, earlier-stage companies that could become acquisition targets or breakout growers if their AI products gain traction. The middle ground — mid-cap software companies with uncertain AI strategies — is being treated with increasing caution.
Hedge fund filings from the first quarter of 2025 showed notable increases in positions in AI-adjacent software names and reductions in traditional SaaS companies with slowing growth profiles. According to Goldman Sachs’ prime brokerage data, net exposure to software within hedge fund portfolios reached its highest level since early 2024, but the composition of that exposure has shifted dramatically toward AI-linked names.
The Road Ahead for Software Investors
The separation between AI winners and losers in software is likely to accelerate in the second half of 2025 and into 2026. Several catalysts could sharpen the divide further. First, the major cloud providers — Amazon Web Services, Microsoft Azure, and Google Cloud — are all expanding their AI service offerings, which will create both opportunities and competitive threats for independent software vendors. Second, enterprise AI budgets are expected to grow by 25-35% in 2025, according to estimates from Gartner, but that spending will be concentrated among vendors that can demonstrate clear return on investment.
Third, and perhaps most consequentially, the emergence of AI agents — autonomous software programs that can execute multi-step tasks without human intervention — threatens to reshape entire software categories. Companies that can build or integrate agent capabilities into their platforms stand to capture significant value. Those that cannot may find their products relegated to commodity status. Salesforce’s Agentforce platform and Microsoft’s Copilot Studio are early examples of this trend, and both companies are betting heavily that agent-based AI will define the next phase of enterprise software.
For investors, the message from the experts is clear: the AI trade in software is not a rising tide that lifts all boats. It is a powerful current that is pulling some companies forward and dragging others under. The discipline to distinguish between the two — using hard revenue data rather than press releases and keynote demos — will be the defining skill for technology investors over the coming years.
WebProNews is an iEntry Publication