By Maksym Misichenko · Nasdaq ·
What AI agents think about this news
The panelists generally agree that agentic AI poses significant risks to ServiceNow (NOW) and Palantir (PLTR), primarily margin compression due to seat-based pricing erosion and potential disruption by hyperscalers offering native agent orchestration. They express bearish sentiments, with valuations being a key concern.
Risk: Margin compression due to seat-based pricing erosion and potential disruption by hyperscalers.
Opportunity: None explicitly stated.
This analysis is generated by the StockScreener pipeline — four leading LLMs (Claude, GPT, Gemini, Grok) receive identical prompts with built-in anti-hallucination guards. Read methodology →
Agentic artificial intelligence (AI) can handle many of the tasks SaaS companies perform.
ServiceNow and Palantir both offer platforms that oversee and manage AI agents.
Both stocks are trading at a hefty premium, even as their stocks go lower.
Will AI create the world's first trillionaire? Our team just released a report on the one little-known company, called an "Indispensable Monopoly" providing the critical technology Nvidia and Intel both need. Continue »
The fear about how agentic artificial intelligence (AI) is making software-as-a-service (SaaS) companies obsolete is real. Many SaaS stocks have tanked over the past year, and these companies are trying to step up their models to remain relevant in an AI world.
The fear has reached SaaS powerhouses ServiceNow (NYSE: NOW) and Palantir Technologies (NASDAQ: PLTR), and their stocks are down 33% and 23% this year, respectively. But agentic AI may not be the threat to these companies that some investors are worried it will be. Here's what you need to know.
SaaS has become the go-to model for software companies, providing recurring revenue through monthly subscriptions. Clients don't get a one-time package that can become obsolete over time but an ongoing relationship with upgrades. SaaS can cover everything from Salesforce's marketing software to Figma's graphic design and Microsoft's Word.
Agentic AI allows the client to automate many of the actions an SaaS program does, making it superfluous. In a case of a subscription that acts as a database, for example, which still needs human interaction to analyze data and take action, the agent can add data, analyze it, and create marketing campaigns based on it.
The SaaS stocks that are hurting the most are the legacy kind. These are companies that have been around for years and are based on older technology. Although everyone is scrambling to keep up with AI, even if these companies switch to an AI profile, their services will be less in demand precisely because their models become more efficient. In other words, they can perform more work faster, needing fewer "seats," or subscriptions.
ServiceNow and Palantir are both AI-based, and their businesses already handle much of the work an agent would. ServiceNow launched its Control Tower last year, before the market started worrying about agentic AI. As the name implies, it's a "centralized command center" to manage a company's AI program.
Even if agentic AI handles many software applications, it still needs to be unified for maximum efficiency and value, and that's where ServiceNow comes in. The company has thousands of clients with long-standing relationships, giving it exposure to further deepen their dependency on its platform.
It's similar for Palantir, which not only unifies disparate datasets but also sends trained engineers to clients to map out its ontology and create a customized system to pull the most actionable insights. Its government and commercial contracts span years, and it's already deeply embedded in their systems. These are advantages that can't simply be replaced by agents and give them a leg up.
As companies rely more on agents, they'll need platforms like Control Tower and Palantir's Artificial Intelligence Platform (AIP) to manage them, creating compelling long-term opportunities.
Both ServiceNow and Palantir are high-growth, profitable companies that are demonstrating strength despite the SaaS stock rout. However, that doesn't necessarily mean they're screaming buys right now, and that's because of one very important factor: valuation. They trade at 61 and 154 times trailing-12-month earnings, respectively, even at their lower prices.
While they may have long growth runways, their upside may be limited by their current price.
Before you buy stock in ServiceNow, consider this:
The Motley Fool Stock Advisor analyst team just identified what they believe are the 10 best stocks for investors to buy now… and ServiceNow wasn’t one of them. The 10 stocks that made the cut could produce monster returns in the coming years.
Consider when Netflix made this list on December 17, 2004... if you invested $1,000 at the time of our recommendation, you’d have $465,733! Or when Nvidia made this list on April 15, 2005... if you invested $1,000 at the time of our recommendation, you’d have $1,313,467!
Now, it’s worth noting Stock Advisor’s total average return is 985% — a market-crushing outperformance compared to 211% for the S&P 500. Don't miss the latest top 10 list, available with Stock Advisor, and join an investing community built by individual investors for individual investors.
**Stock Advisor returns as of May 30, 2026. *
Jennifer Saibil has no position in any of the stocks mentioned. The Motley Fool has positions in and recommends Figma, Microsoft, Palantir Technologies, Salesforce, and ServiceNow. The Motley Fool has a disclosure policy.
The views and opinions expressed herein are the views and opinions of the author and do not necessarily reflect those of Nasdaq, Inc.
Four leading AI models discuss this article
"61x and 154x earnings leave no margin for the execution or competitive risks that agent orchestration platforms still face."
The article correctly notes ServiceNow's Control Tower and Palantir's AIP could manage fleets of agents, leveraging existing enterprise relationships. However, it glosses over integration friction and the risk that hyperscalers (MSFT, GOOG) embed agent orchestration directly into their clouds, eroding the need for separate platforms. NOW and PLTR already trade at 61x and 154x trailing earnings after 23-33% YTD declines; any delay in agent-driven revenue or margin pressure from custom ontology work would trigger further compression. Government contracts help Palantir but expose it to budget cycles and competition from classified alternatives.
Their multi-year, deeply embedded contracts and data ontologies create switching costs that pure agent frameworks cannot replicate quickly, potentially sustaining premium multiples if adoption accelerates faster than expected.
"The article mistakes a real competitive threat (agentic AI) for a moat, when the actual risk is that NOW and PLTR become middleware in a hyperscaler-dominated stack—valuable but not trillion-dollar-platform valuable."
The article conflates two separate problems. Yes, agentic AI threatens narrow SaaS tools—but the core claim that NOW and PLTR become 'trillion-dollar platforms' rests on unproven assumptions: (1) that AI agent management becomes a bottleneck valuable enough to sustain 61x and 154x P/E multiples, and (2) that switching costs and embedded relationships prevent disruption. The article ignores that if agents truly automate workflows end-to-end, the *need* for centralized management platforms may shrink, not grow. Control Tower and AIP are valuable, but they're not immune to commoditization or to being bundled into cloud providers' native offerings. Valuation is the real tell: these multiples price in near-perfect execution for years.
If agentic AI actually works as promised, it will likely be bundled into AWS, Azure, or Google Cloud at near-zero marginal cost, and NOW/PLTR's 'management layer' becomes a feature, not a platform. Switching costs matter less when the alternative is free.
"ServiceNow and Palantir face a valuation cliff if they cannot successfully transition from seat-based SaaS revenue to value-based pricing in an agent-driven ecosystem."
The article frames ServiceNow (NOW) and Palantir (PLTR) as 'agent-proof' platforms, but it misses the fundamental risk of margin compression. If agentic AI truly automates workflows, customers will demand lower seat-based pricing, threatening the high-margin subscription models that justify these 60x-150x P/E multiples. While NOW and PLTR provide the 'control plane' for agents, they are essentially betting that they can capture the value of the automation they enable. If they fail to pivot their billing from 'per-seat' to 'per-outcome' or 'per-agent-transaction,' they risk becoming utilities rather than premium platforms. I am skeptical that their current valuations account for the inevitable cannibalization of legacy SaaS revenue.
The counter-argument is that these platforms are becoming the 'operating system' of the enterprise, and as agent complexity scales, the value of the central nervous system (the platform) increases, granting them massive pricing power that offsets any per-seat decline.
"AI-agent hype must translate into durable earnings power for NOW and PLTR, or risk a re-rating as moat durability comes into question."
Reading this, the thrust is: agentic AI will solidify NOW and PLTR as enduring platform moats, not obsolete SaaS. The logic hinges on governance, data unification, and custom human mapping that only incumbents provide. Yet the piece glosses over two risks: first, a spike in valuation multiples for growth plays may not justify profits if AI adoption compresses price per seat or shifts revenue to services. Second, rival platforms from hyperscalers and new AI orchestration layers could undermine lock-in by enabling customers to mix and match tools, shrinking the incremental revenue from core platforms. Palantir's and ServiceNow's government exposure and integration costs add further sensitivity.
Strongest counter: even if agents threaten routine SaaS seats, enterprises may double down on NOW/PLTR governance and data networks, creating sticky, cross-sell-rich platforms. That could support higher multiples if execution milestones prove durable.
"Hyperscaler bundling threatens valuations faster than pricing model shifts."
Gemini's focus on per-seat erosion misses how outcome pricing could lift total revenue if agents multiply workflows 5-10x, yet this hinges on NOW/PLTR locking in the ontology layer before hyperscalers absorb it. A clearer risk is co-opetition: MSFT could embed PLTR-like governance into Fabric or Dynamics at marginal cost, converting high-margin platform deals into lower-value services and accelerating multiple compression beyond what seat cannibalization alone implies.
"Bundling risk assumes hyperscalers can replicate enterprise data governance at scale, which is unproven and may be harder than building a platform from scratch."
Claude and Grok both assume hyperscalers will embed orchestration 'at marginal cost,' but that's underspecified. Azure's agent layer still requires enterprise data governance, compliance mapping, and custom ontologies—exactly what NOW/PLTR own. Hyperscalers are distribution, not ontology engines. The real question: can MSFT/GOOG build enterprise data unification faster than NOW/PLTR can lock in customers? If not, bundling threat is overstated.
"Hyperscaler bundling creates a 'good enough' alternative that forces NOW and PLTR into a margin-eroding battle against free, native infrastructure."
Claude, you undervalue the 'distribution as a feature' trap. Hyperscalers don't need to build a perfect ontology engine; they just need to build a 'good enough' one that is free, integrated, and compliant. Once Azure or GCP offers native, secure agent orchestration, the friction of moving data to a third-party platform like PLTR becomes a massive hurdle for CIOs. NOW and PLTR are fighting a war against free, bundled infrastructure, and that is a margin-crushing battle they cannot win long-term.
"Valuation depends on durable differentiation in governance and ontology services, not just the threat of hyperscaler bundling."
Responding to Gemini: Even if hyperscalers offer 'good enough' native orchestration, the real moat for NOW/PLTR is enterprise data governance and bespoke ontology work that yields auditable, cross-organization controls. Bundling at marginal cost reduces price power, but customers won't drop governance requirements overnight. The risk isn't only seat pricing; it's potential compression of margins if 'per-ontology' or 'per-control' services become standardized. Valuation hinges on durable differentiation, not just avoidance of disruption.
The panelists generally agree that agentic AI poses significant risks to ServiceNow (NOW) and Palantir (PLTR), primarily margin compression due to seat-based pricing erosion and potential disruption by hyperscalers offering native agent orchestration. They express bearish sentiments, with valuations being a key concern.
None explicitly stated.
Margin compression due to seat-based pricing erosion and potential disruption by hyperscalers.