2 AI Stocks With 85% and 70% Upside to Buy During a Software Bear Market

The article argues ServiceNow and Salesforce are well positioned as AI orchestration and agent-launch platforms, with estimated upside of 85% for ServiceNow and 70% for Salesforce based on valuation assumptions. ServiceNow is cited as growing revenue around 20% and Salesforce as targeting about 11% CAGR through fiscal 2030, supported by Data 360 and Informatica. The piece is mostly bullish commentary rather than new company-reported results, so likely market impact is limited.

Analysis

The market is still pricing SaaS as if agentic AI is a direct substitute for software, but the more likely near-term outcome is a redistribution of value toward systems that own permissions, governance, and workflow state. That favors platform vendors with deep entrenchment and penalizes thin UI layers, but the bigger second-order effect is on implementation budgets: if customers shift spend from seat licenses to orchestration, governance, and data-quality layers, the software stack becomes more concentrated around a few control points rather than disintermediated.

NOW looks better insulated than the market implies because AI amplifies the value of being the transaction and approval layer, not just the screen. If AI agents become the primary users, the vendor that can enforce auditability, role-based access, and exception handling should gain share versus point tools; the risk is not commoditization but slower expansion if customers pause large transformation projects while they re-architect around AI. CRM has a similar advantage, but its upside is more dependent on proving that data unification and governance convert into monetizable workflow throughput rather than just being a cleaner data lake.

The contrarian miss is that AI could be bullish for a subset of SaaS because it raises the cost of building and maintaining custom internal tools. Many firms will try to automate internally, but the failure mode is operational debt, security exposure, and brittle maintenance, which usually pushes enterprise buyers back toward vendors with compliance and support. That makes the timeline important: over the next 6-18 months, the first winners should be the vendors that can sell “safe AI operations” rather than generic copilots.

INFA is the quieter beneficiary in this stack because data quality and governance become the bottleneck once zero-copy and multi-cloud ingestion reduce the friction of moving data. PLTR remains a secondary winner, but its upside is more policy- and defense-driven than a pure SaaS rerating; BABA/NVDA/INTC are only tangentially affected here, mainly through the model/infrastructure layer rather than application monetization.

Key Decisions for Investors

• Go long NOW vs a basket of lower-quality SaaS UI wrappers over the next 3-6 months; the setup favors companies with embedded workflow control and should outperform if AI spend shifts toward orchestration.
• Initiate a long CRM / short a basket of generic SaaS names pair trade; target a 6-12 month horizon where data-governance monetization can re-rate CRM while seat-based apps face multiple compression.
• Buy INFA on weakness as a second-order AI governance beneficiary; the best risk/reward is 6-12 months out if enterprise data cleanup becomes the gating item for agent deployment.
• Use short-dated put spreads on vulnerable SaaS names that rely on UI abstraction rather than system-of-record depth; catalyst window is the next earnings season if management teams fail to show AI-driven net retention expansion.
• Avoid chasing PLTR/NVDA here on this thesis alone; their upside is real but the article’s edge is in application-layer control points, not infrastructure beta.