Forget SaaS: Why AI Agents Could Make ServiceNow and Palantir Technologies the Next Trillion-Dollar Platforms

ServiceNow and Palantir are down 33% and 23% this year, respectively, as investors worry that agentic AI could make SaaS software less necessary. The article argues both companies are relatively well positioned because their platforms already manage AI workflows, but it cautions that valuations remain rich at 61x and 154x trailing earnings. Overall, the piece is more analytical than event-driven and is unlikely to move the broader market materially.

Analysis

The market is treating “agentic AI” as a binary threat to SaaS, but the more important second-order effect is software spend consolidation: agents reduce user seats while increasing demand for orchestration, identity, audit, and policy layers. That shifts budget from horizontal workflow apps toward control-plane vendors with deep integration, which is why NOW and PLTR can outperform legacy SaaS even if overall seat growth decelerates. The real competitive losers are mid-tier application vendors without proprietary data or embedded workflows; their products become easier to replicate and harder to price on a per-user basis.

The contrarian point is that the best AI names are now being priced like scarce infrastructure, not software, so the upside from narrative expansion is largely gone unless revenue growth re-accelerates meaningfully. At these multiples, any sign that AI modules are cannibalizing core subscription counts faster than they expand ACV would compress valuation quickly over a 3-6 month horizon. Conversely, if AI agents drive larger enterprise deployments, the benefits should show up first in net retention and implementation attach rates, not top-line inflections.

For catalysts, the next 1-2 quarters matter more than the multi-year thesis: watch for commentary on seat optimization, AI module penetration, and whether customers are standardizing on a few orchestration platforms instead of experimenting broadly. A risk to the bullish thesis is that hyperscalers and workflow incumbents bundle similar control features into broader platforms, reducing pricing power for standalone vendors. The cleaner trade is not “AI software long,” but “winner-take-most control layer long versus legacy workflow short,” because the market is still underestimating the value of governance in an agent-heavy enterprise stack.

Key Decisions for Investors

• Maintain a tactical long bias in NOW for the next 1-2 quarters, but size it as a quality compounder trade rather than a momentum chase; upside is in multiple durability if Control Tower adoption offsets seat compression, while downside is a 15-20% de-rating if growth slows.
• Prefer PLTR over broad SaaS exposure for a 6-12 month horizon: it has the stronger data/implementation moat and should capture more budget as enterprises standardize agent governance; use pullbacks to initiate, but avoid paying peak multiple expansion.
• Pair trade: long NOW / short a basket of legacy workflow SaaS (CRM, FIG) for 3-6 months; thesis is that agent adoption accelerates value migration to control planes while seat-based software with weaker data moats faces the steepest pricing pressure.
• If already long NOW or PLTR, hedge valuation risk with call spreads rather than outright longs; the stocks can work operationally while still underperforming on multiple compression if the market rotates away from premium software.
• Set a trigger to fade both names on any quarter where AI module growth rises but gross seat count or subscription count decelerates materially; that would confirm the market’s cannibalization concern and likely cap upside for 1-2 quarters.