Enterprise AI: Value Capture at Scale
Scaling AI across industrial operations is not just about deploying more models. It requires a new digital ecosystem that can connect data, engineering discipline, operational context and enterprise workflows at scale. In this presentation, Michel Teughels of ExxonMobil explores how organisations can move from isolated AI capability towards enterprise-wide value capture.
Using the example of operating a heat exchanger, the session highlights the complexity of applying digital tools across real industrial environments. Multiple personas, systems, models, constraints and workflows all interact around a single asset, creating integration challenges that can limit the pace at which AI and Gen AI can be deployed and maintained.
The presentation positions AI as an augmentation layer rather than a replacement for engineering understanding. By combining first-principles models with AI, organisations can retain rigorous asset knowledge while using intelligent tools to handle complexity, scale insights across fleets and support faster, more resilient operations.
A key theme is the need for an open, interoperable digital ecosystem. Michel explores how knowledge graphs, shared semantics, reality capture, engineering data, real-time data and open standards can reduce integration friction, prevent lock-in and make operational data more reusable across assets, processes and use cases.
The presentation concludes with a call for industry collaboration, using AI-augmented HAZOP as a practical proof of value. Read through the full presentation to see how ExxonMobil is framing enterprise AI, open digital ecosystems and reusable data foundations as critical enablers for value capture at scale.
