Industry AI: Deep Industry Knowledge, Forward-Deployed Engineering, Measurable Value
Enterprise AI has reached an inflection point.
But while there is enormous excitement about what AI can do, we are still measuring progress the wrong way. The number of models, pilots, or agents a company deploys tells us very little. What matters is whether AI transforms the core processes where business value is actually created.
Can it steer demand fulfillment to high-margin business when supply is disrupted? Can it help release a pharmaceutical batch without compromising quality and compliance? Can it resolve a critical maintenance issue before it turns into an outage?
This is where enterprise AI becomes real and where generic intelligence reaches its limits.
From pharmaceutical supply chains and manufacturing quality management to asset maintenance, trading, and retail, the hardest industry problems are deeply contextual. They are shaped by decades of interconnected processes, strict regulations, and operational constraints. Solving them demands more than a powerful model. It requires trusted business data, integrated processes, rock-solid governance, and a deep industry domain expertise–the very foundation upon which we are building SAP’s Industry AI. At the same time it demands a team which is obsessed with customer problem solving and world-class forward deployed engineers.
This convergence of industry depth and forward deployed engineering (FDE) is where the next wave of enterprise value will be unlocked.
Start with the value, not the technology
AI innovation too often turns into a search for problems to solve. We take the opposite approach: start with the customer’s desired business value and work backward.
Where is margin leaking? Why is there lost sales? What operational decisions take too long? What production deferment can be prevented?
The goal is no longer just incremental productivity for individual users. It is moving toward autonomous operations. Across industries, organizations need to compress the cycle between detecting an issue, simulating options and impact, deciding the right response, executing corrective actions, and learning from outcomes–in a compounding loop. What once took days can increasingly happen in minutes.
Through our forward-deployed engineering model, we co-develop tailored capabilities alongside customers. We solve high-stakes challenges that have no off-the-shelf answers, then turn those proven solutions into scalable software capabilities for entire industries. This is not innovation from a distance. It is built with the people who live these challenges every day.
Autonomous operations make this possible. Within clearly defined business guardrails, AI agents identify disruptions, model scenarios, evaluate trade-offs, orchestrate execution, and improve via feedback loops–with human experts firmly in the loop and on-the-loop to validate critical decisions and steer strategic direction.
Co-innovation as a catalyst for impact
The most valuable AI solutions are never built in isolation. They are forged inside live operating environments.
Takeda Pharmaceuticals, a global leader in research-based biopharmaceuticals, is applying this thinking to its future operating model.
According to Rebecca Kaufmann, senior vice president and head of Enterprise Platforms a Takeda Pharmaceuticals, “AI represents the key enabler for Takeda’s transformation for the future. We are combining the SAP industry and technology knowledge with our organizational real-life experience. The supply chain planning assistant will provide us with AI-detected root causes and AI-suggested remediation and also automated performance tracking resulting in a self-healing capability.”
That ambition–shifting from passive detection to systems that continuously learn, recommend, and self-heal—is taking hold across sectors.
Doehler, a global producer of natural ingredients for the food, beverage, life sciences, and nutrition industries, is already demonstrating what this looks like on the factory and logistics floor.
“Our people’s greatest value lies in deciding what to do next, and now they can focus entirely on that,” said Stephan Schissler, co-chief information officer at Doehler GmbH. “The Logistics Issue Resolution Agent detects the root cause, recommends the resolution, and turns a multi-hour firefight into a one-click decision. Our logistics team has recovered roughly a third of the time once lost to troubleshooting, and our processes are becoming self-resolving rather than exception-driven.”
These examples reinforce a decisive rule: AI delivers maximum return when it eliminates operational drag, freeing professionals to focus on judgment, strategy, and business outcomes.
AI that thinks like the industry
As AI moves from passive recommendation to intelligent execution, industry context becomes paramount.
The question is no longer whether an AI agent can take an action. The question is whether it understands which action is compliant, safe, and commercially viable within a specific environment and organizational guardrails.
- In life sciences, autonomous manufacturing coordinates processes from design-to-release while strictly adhering to compliance-critical quality and compliance criteria.
- In energy and natural resources, agents go beyond detecting maintenance alerts to dynamically schedule crews, stage spare parts, and rebalance operations.
- In engineer-to-order manufacturing, systems orchestrate complex quotations, adapt production schedules, and accelerate new product introductions.
Deep industry context is what makes these outcomes trustworthy.
Building intelligence that compounds
SAP’s Industry AI portfolio is focused across seven core domains: Autonomous Asset Management, Autonomous Commodity Management, Autonomous Adaptive Production, Autonomous Regulated Manufacturing, Autonomous Revenue Growth Management, Autonomous Unified Commerce, and Autonomous Project Delivery.
While the use cases differ, the principle is the same: sustainable value comes from end-to-end value streams, not optimizing functional silos.
“Business performance depends on how quickly a company can respond when conditions change,” said Andre Bechtold, president of SAP Industries & Experiences and chief revenue officer of Industry AI at SAP. “By applying SAP’s industry knowledge across connected processes, Industry AI helps customers act earlier, limit disruption, and keep operations on track.”
The measure that matters
Coming out of SAP Connect in 2026, one message from our customer discussions was unmistakable: the ambition across sectors is to move past fragmented experimentation and embed agentic intelligence directly into mission-critical workflows.
Our initial wave of early adopter customers has proven the compounding power of this model: Customer friction points sharpen our agentic solutions, stronger agentic solutions strengthen standard, and stronger standards accelerate the time to value the next customer can achieve.
The future of enterprise AI will not be decided by who launches the most agents. It will be decided by who can bring intelligence into the decisions and workflows where business value is created every day.
When AI understands the industry, operates within trusted boundaries, and learns alongside human experts, autonomy stops being an abstract vision. It becomes a better way to run the business.
For a complete overview of the Industry AI innovations and product announcements unveiled at SAP Connect in 2026, see the SAP Connect Innovation News Guide and related SAP News coverage.
Dominik Metzger is president and global head of Industry AI at SAP.
