Tech Disruptors: SAP’s COO on Why AI Needs Better Foundations

SAP COO Sebastian Steinhaeuser said AI is a major technology shift, but emphasized that customers still need strong foundations in cloud, data modernization, and application consolidation before deploying agents and assistants. The discussion highlighted SAP's autonomous enterprise vision and its strategy to embed business-process context, governance, and industry-specific knowledge into its AI layer. The piece is largely strategic commentary with limited immediate market-moving information.

Analysis

The key equity implication is that AI monetization is likely to be a multi-year process layered on top of a much more immediate migration cycle: the firms that can bundle process software, data governance, and workflow automation should see AI act as a retention tool rather than a standalone revenue line. That favors incumbent enterprise platforms with deep installed bases and high switching costs, while pure-play AI application vendors face a higher burden of proof on ROI and integration. In practice, the market may be underpricing how AI increases the value of boring infrastructure spend, not just experimental pilot budgets.

The second-order effect is pressure on software pricing architecture. If customers demand measurable business outcomes, vendors will need to shift toward consumption, outcome-based, or agent-seat pricing, which compresses near-term visibility but can expand lifetime value for winners with strong usage density. That creates a near-term multiple risk for names perceived as “AI beneficiaries” without clear workflow control, while increasing the strategic value of platforms that sit between data, ERP, and governance layers.

A useful contrarian read is that AI here is less a discretionary acceleration story and more a validation of cloud and application consolidation. If AI agents need clean master data and standard processes, then fragmented stacks become a liability; that can accelerate consolidation budgets even in slower IT spending environments. The risk is timing: adoption may remain pilot-heavy for 2-3 quarters, so shares can rerate before earnings fully inflect, but the actual revenue payoff could lag until customers prove the productivity gains internally.

Near-term reversal risk is mostly execution-related: if agentic features fail to show hard savings, or if model commoditization makes application-layer differentiation thinner than expected, enthusiasm for “AI-enabled enterprise software” could fade quickly. Longer term, the more dangerous risk for incumbents is disintermediation by AI-native workflow tools that sit above the ERP layer, but that requires solving trust, auditability, and industry-specific context — a much higher bar than consumer AI.

Key Decisions for Investors

• Long SAP vs. short a basket of AI-native enterprise workflow names with limited installed base, 3-6 month horizon. The thesis is that governed process control monetizes faster than standalone AI features; stop if SAP loses evidence of net retention improvement.
• Buy SAP on pullbacks for a 12-month hold, targeting a 10-15% rerating as AI becomes a retention and upsell lever rather than a cost center. Risk/reward is favorable if cloud migration re-accelerates; trim if pricing shifts begin to pressure ARR visibility.
• Pair trade: long enterprise infrastructure/cloud enablers with governance exposure, short horizontal application software lacking workflow ownership, 6-9 months. This captures the likely reallocation toward foundational spend as AI adoption moves from demo to production.
• Consider SAP call spreads 6-12 months out to express upside from agent adoption with defined premium outlay. The trade benefits if market starts capitalizing AI-driven cross-sell before reported revenue fully catches up.
• Set a catalyst watch on next 1-2 earnings cycles for indicators of higher consumption or attach rates tied to AI features. If there is no measurable uplift by then, reduce exposure because the market may have already priced the strategic narrative.