How enterprise data becomes an innovation accelerant

Key Concepts

• Change-Readiness: The ability of an organization to adapt to and thrive amidst rapid change, balancing AI and human elements.
• Experimentation: A core strategy for competitive advantage in a fast-paced environment, allowing for testing, learning, and faster iteration.
• Embedded AI: Artificial intelligence integrated directly into business processes and applications, providing contextualized insights and actions.
• SAP Business Suite: SAP's integrated suite of applications designed to manage end-to-end business processes, now enhanced with AI capabilities.
• Deterministic vs. Probabilistic Technologies: Deterministic systems provide precise, predictable outcomes, while probabilistic systems (like generative AI) offer insights based on data patterns, with potential for variability.
• Contextualization: The ability of AI to understand and apply information within a specific business context, crucial for enterprise-grade AI.
• Semantic Search: A search method that understands the meaning and context of queries, rather than just matching keywords, leading to more accurate results.
• Agentic AI: AI that can perform tasks and make decisions on behalf of users, acting as an interface to systems.
• Virtuous Cycle: A continuous loop where better data leads to better AI, which drives more usage and thus more data, creating a self-reinforcing cycle of improvement.
• Data Products: Bundles of infrastructure, AI code, insights, and sometimes dashboards, developed by SAP and its ecosystem, that can be integrated with various data sources.
• Augmentation vs. Replacement: The perspective that AI should enhance human capabilities rather than solely replace human roles.

Experimentation as a Competitive Edge

Manos Raptopoulos emphasizes that in today's rapidly changing world, "standing still is moving backward." He argues that experimentation is the new competitive edge. Companies are constantly seeking to accelerate their progress and counter competitive threats. SAP Business Suite facilitates this by enabling companies to test, learn, and move faster than their competitors. This iterative process, which starts with a business issue and feeds into product development, has always been the foundation of SAP's solutions, evolving from integrated systems and end-to-end process management to incorporating AI.

Regional Perspectives and Challenges

While Jan Gilg provided an Americas perspective, Manos Raptopoulos offers insights from Europe, Asia, and the Middle East. He highlights the interconnectedness of global markets, where events in one region can have a domino effect elsewhere, citing the impact of geopolitical events like tariff wars.

• AI Landscape: The U.S. has a strong AI tech sector with large companies. Europe has a thriving ecosystem but fewer dominant tech giants. SAP plays a role in both innovation and regulation, with Europe having a more stringent regulatory environment compared to the U.S.
• China's Role: China is a significant player in every technological element, including AI and generative AI, possessing a large economy with considerable gravitational force.
• Middle East's Approach: The Middle East demonstrates a propensity to innovate and "leapfrog" traditional development stages, contrasting with Europe's focus on business cases, bottom-line impact, and ROI.

Navigating AI Hype and Grounding in Business Outcomes

Manos Raptopoulos expresses passion for grounding AI in tangible enterprise business outcomes. He acknowledges that AI is not new but has evolved, with generative AI marking a significant breakthrough. However, he stresses the importance of getting AI right in an enterprise context.

• Data Security and Authorization: A critical concern is ensuring generative AI agents do not provide unauthorized information. Unlike the open internet, enterprise data requires strict access controls.
• Deterministic vs. Probabilistic Outcomes: Raptopoulos highlights the critical difference between deterministic technologies (which provide exact answers) and probabilistic technologies like generative AI. In business, precision is paramount. For instance, a slight difference in EBITDA (e.g., 15.7% vs. 15.9%) can significantly impact analyst expectations and valuations. CFOs require deterministic certainty.
• SAP's Role: SAP leverages its access to high-quality data from a multitude of global industries to train models that can achieve more deterministic outcomes, addressing the "last mile" challenge of AI implementation.

Joule: Embedded AI within SAP Business Suite

Joule is described as embedded AI within the SAP Business Suite, with capabilities ranging from invoice processing to applicant evaluation and supply chain disruption prediction.

• Seamless Integration: Joule's integration across departments is attributed to SAP's robust technology stack and the continuous development of Joule's skills.
• Enterprise-Grade Capabilities: SAP focuses on building enterprise-grade AI that provides accurate answers within a business context.
• Contextualization as a Differentiator: Joule's key differentiator is its ability to contextualize information.
• Single Entry Point: Joule acts as a single entry point into SAP's environment, allowing users to ask any question and be guided to the relevant data and functionalities within the Business Suite, regardless of department.
• Extending Beyond SAP: Joule is being developed to extend beyond the SAP context. For example, an invoice, a business object within SAP, is connected to contracts, payment terms, quality management, and payment systems. Extracting this invoice to an external data source for a third-party model would lose these semantic connections, requiring costly recreation. Joule preserves this context, making it "business AI relevant" and a more cost-effective solution.
• Semantic Search Power: The transcript contrasts keyword-based search with semantic search, which understands context and meaning, making it a more intelligent and effective way to find information.
• New User Interface Paradigm: SAP envisions Joule becoming the new user interface, moving beyond traditional menu-driven interactions. Agentic AI can perform tasks, present solutions, and act on them based on programmed criteria or user direction, creating a more fluid and democratic interaction with systems.

The End of Dashboards and Hybrid Approaches

Raptopoulos predicts the end of dashboards, with insights appearing directly within workflows when needed. However, he acknowledges that enterprises may initially cling to dashboards as a safety net.

• Hybrid World: A hybrid approach is expected for a period, where users will trust familiar systems while gradually adopting new AI-driven insights.
• User-Centricity and Trust: Change management and user trust are crucial for technology adoption. The transition requires users to be ready to let go of old methods.
• Staged Approach: SAP is ensuring it continues to offer robust dashboarding and data visualization capabilities while simultaneously enhancing Joule and generative AI functionalities. This staged approach is vital for building user trust, likened to a parent allowing a teenager to drive with them before handing over the car keys.

The Virtuous Cycle of Data, Applications, and AI

SAP's Business Suite is designed to deliver a virtuous cycle where applications, data, and AI continuously feed into one another, enabling constant learning, prediction, and adjustment.

• Data Quality and AI Improvement: The principle is that better data leads to better AI algorithms, which in turn encourages more user engagement and data input into the ecosystem.
• Ecosystem Role: SAP believes its existing and future ecosystem plays a significant role in this cycle. Partners are developing applications on the SAP framework, creating an "application store" of data products. These data products combine infrastructure, AI code, insights, and dashboards, and can be coded into by anyone, regardless of the data source.
• Snowball Analogy: This cycle is likened to a snowball rolling downhill, growing larger and more powerful.

AI, Innovation, and the Future of Work

The discussion delves into the deeper philosophical question of whether AI agents will steer innovation and if this diminishes its virtue.

• Job Creation vs. Destruction: Acknowledging the broader societal question of AI's impact on jobs.
• AI as an Accelerator: The focus is on leveraging AI as an accelerator and tool, rather than giving away too much control that could hinder future innovation.
• Ethical AI Development: Technology companies and individuals must develop AI ethically, with humanity at the center.
• Societal Preparation: Society needs to prepare for AI's impact through training, reskilling, and understanding. Control should be maintained, avoiding the premature handover of critical responsibilities.
• Augmentation as the Goal: The preferred perspective is that AI should augment human capabilities rather than replace them.

Conclusion

Manos Raptopoulos concludes that adaptability is today's advantage, and resilience alone is no longer sufficient. The key to future business success lies in foresight, flexibility, and forward motion, driven by the integration of experimentation, adaptability, and embedded AI within solutions like SAP Business Suite.