2026 and the AI Flywheel

2026 is the year when the business value of AI will really take off.

It’s no longer only about AI filling frustrating gaps between existing systems, it’s about AI that “just works”, seamlessly embedded in existing workflows, accelerating every aspect of the business.

It’s never been easier to spin up AI prototypes, but it’s never been harder to put them into production. GenAI is amazingly flexible, but it’s also amazingly flexible with its answers, and so needs more robust frameworks, grounding, and guardrails than most organizations realize.

The first generation of AI in business was mostly hand-built, like the first automobiles. But now we’ve moved to modern assembly lines, designed to make it easier and faster to create new business AI use cases, with higher quality.

It takes a while to spin up an enterprise-business flywheel, but now we’re seeing the acceleration of implemented AI business cases, all underpinned by a enterprise-ready AI foundation built on SAP Business Technology Platform.

It provides everything you need to build, run, and integrate AI agents and custom AI solutions at scale, including orchestration, knowledge graph, prompt optimization, fine-tuning, data anonymization, large choice of models, output filters and more. It’s robust because it’s the platform that SAP uses to provide the mission-critical AI directly in the Business Suite.

Check out the latest version of the SAP Innovation Guide, now updated after this week’s SAP TechEd event — I guarantee you’ll see things you didn’t realize were now possible!