Logic Behind ServiceNow and Palantir's Trillion-Dollar Valuation in the Age of AI Agents
0xBroomberg
AI agents are disrupting traditional SaaS; ServiceNow and Palantir are down 33% and 23% this year — yet both are built on AI platforms that could make them winners, not victims, of the shakeout.
Who is AI actually disrupting?
AI agents — programs that automate data entry, analysis, and campaign planning — are replacing the tasks that traditional SaaS tools charged for on a per-seat basis.
This means → the hardest hit are legacy SaaS companies still running on old architectures and selling subscriptions by headcount. The per-seat model is breaking down.
ServiceNow and Palantir are already built on AI. They are not the tools being replaced — they sit at the platform management layer that orchestrates the agents.
Why can ServiceNow survive this wave?
Last year it launched Control Tower, positioning it as a centralized command center for all of a company's AI programs.
In plain terms = even if AI agents take over individual software tasks, enterprises still need one unified platform to manage and coordinate them — that is the role ServiceNow is claiming.
It has long-term relationships with thousands of enterprise clients. That deep integration makes migration costly, and this reflects how the management-layer position itself becomes a moat.
Where is Palantir's moat?
Palantir's AI Platform, AIP — a system that pulls scattered enterprise data into one place — goes beyond selling software: it embeds engineers on-site at client facilities to build customized data architectures.
Government and commercial contracts run long, and the system is deeply embedded. This means → once a client adopts it, switching costs are extremely high, and AI agents cannot easily bypass it.
Put simply = Palantir is not selling an app you can uninstall. It is selling a data nervous system that grows inside the client's operations.
At these valuations, is there still upside?
On a trailing-twelve-month basis, ServiceNow trades at roughly 61× earnings; Palantir at roughly 154×.
Both companies are genuinely high-growth and profitable, but this reflects a stock price that has already priced in a large share of future expectations.
In plain terms = the long-term growth story may be real, but near-term upside is compressed by the premium already baked in — even a great company delivers weaker returns when you overpay.
Content is for reference only, not financial advice.