KnowPriors• Apr 23, 2026• 39:48Interview

SAP: Bringing the ‘Operating System’ of a Company into the AI Era with CTO Philipp Herzig

From KnowPriors

Philipp Herzig•CTO, SAP

Executive Summary

SAP is undergoing a significant transformation, re-engineering its entire software stack around AI, with a focus on agentic systems that move beyond traditional user interfaces.
The company is leveraging AI to drive productivity, citing a 30% effort reduction for consultants using its "Joule for Consulting" product, and is applying agentic coding across all internal development.
AI is catalyzing a fundamental business model shift for SAP, moving away from traditional seat-based licensing towards consumption-based pricing, though currently managed through a hybrid model to meet customer needs for predictability.
SAP argues that while Large Language Models (LLMs) excel with unstructured data, specialized models are required for predictive tasks on structured enterprise data, leading them to develop proprietary models like RPT1.

12 quotes

Concerns Raised

The difficulty of ensuring AI agents perform reliably and verifiably at enterprise scale.
Scaling tool use for AI agents across a platform with over 20,000 APIs.
Customer readiness and fear of runaway costs are slowing the transition to pure consumption-based pricing.
Security vulnerabilities in open-source AI tools remain a significant barrier to enterprise adoption.

Opportunities Identified

Reducing consultant effort on complex projects by 30% using AI tools like 'Joule for Consulting'.
Automating mundane tasks in functions like finance and HR to up-level employee roles.
Creating a data flywheel by capturing user context and decisions through 'agent mining' to continuously improve processes.
Replacing static dashboards with conversational analytics, allowing users to query data in natural language.

Key Themes

The Agentic Enterprise

The discussion centers on a paradigm shift from static, human-operated user interfaces to proactive AI agents that automate complex business processes. This involves not just building models, but teaching them to use tools (APIs) reliably and verifiably at massive scale.

This theme highlights the next evolution of enterprise software, where value is derived from automated outcomes rather than simply providing tools for human operators, fundamentally changing how work is done.

AI as a Business Model Catalyst

AI is not just a technological shift but a commercial one, forcing a transition from predictable seat-based licenses to value-aligned consumption-based pricing. The speaker notes this is a journey, with hybrid models serving as a bridge to manage customer concerns about cost predictability.

This signals a major disruption in the SaaS business model. Companies must navigate customer readiness and develop new pricing strategies that capture the value of AI-driven outcomes.

Bridging the AI Innovation-Outcome Gap

The speaker observes a growing disconnect between the rapid pace of AI innovation and its tangible adoption and impact within enterprises. SAP's strategy is to focus on delivering concrete business outcomes, like cost reduction and faster project delivery, to close this gap.

This addresses the core challenge for enterprise AI adoption. Success depends less on having the most advanced model and more on integrating AI to solve real-world business problems and demonstrate clear ROI.

The Engineering Reality of Enterprise AI

The primary engineering challenge is not model creation, but ensuring agents perform tasks correctly and reliably. This has led to a renewed focus on methodologies like Test-Driven Development (TDD) and new concepts like "agent mining" to capture decision context and improve verifiability.

This provides a practical perspective on AI implementation, emphasizing that traditional software engineering discipline, adapted for AI, is crucial for building robust, maintainable, and trustworthy enterprise-grade systems.

Specialized vs. General AI in Business

A key assertion is that general-purpose LLMs are insufficient for core business predictive tasks like demand forecasting. SAP is developing specialized models (e.g., RPT1) designed for structured, tabular data to achieve the necessary accuracy and reliability for these critical functions.

This challenges the notion that a single, large model can solve all problems. It underscores the need for a diverse AI toolkit, where specialized models are deployed for high-stakes, domain-specific tasks.

Get started free

Topics

Enterprise AI AI Agents Agentic Software SAP Business Model Transition Consumption-Based Pricing SaaS Test-Driven Development (TDD)Tool Use API Scaling Process Mining Agent Mining Knowledge Graph Structured Data Predictive AI Large Language Models (LLMs)

Processed Apr 23, 2026 yt-dlp + mlx-whisper + Gemini