Teaching AI to run with the turbines

2026-07-02 · Source: MIT Technology Review · Field: Energy & Utilities — Artificial Intelligence & Machine Learning, Traditional Energy & Fossil Fuels, Utilities & Infrastructure · Depth: Intermediate, extended

Summary

Woodside Energy, a global energy producer, has spent over a decade integrating AI into its asset-intensive, safety-critical operations, moving beyond consumer-facing tools. The company's long-term investment in predictive analytics, optimization systems, and machine learning since 2015, coupled with robust data governance, now supports advanced agentic AI. Key applications include the "Startup Advisor" copilot for managing complex liquefied natural gas (LNG) plant startups and "maintenance intelligence," which optimizes maintenance activities, potentially reducing hours by up to 15% over five years on piloted assets. Woodside's strategy, in partnership with Infosys, emphasizes scaling enterprise-wide AI solutions from small prototypes, supported by standardized platforms and a strong governance framework including structured assessments and an AI council.

Key takeaway

For Directors of AI/ML or VPs of Engineering aiming to scale industrial AI, Woodside Energy's journey highlights the necessity of long-term investment in data foundations and governance before pursuing advanced agentic systems. Your strategy should prioritize enterprise-wide platforms and structured risk assessments, moving beyond isolated solutions. Focus on augmenting human expertise in high-stakes environments, like optimizing plant startups or maintenance, to achieve tangible operational improvements and cost reductions.

Key insights

Industrial AI success hinges on foundational data, robust governance, and augmenting human expertise in critical operations.

Principles

• Think big, prototype small, scale fast.
• Prioritize enterprise-wide, coordinated AI capabilities.
• Strong governance is critical for rapid, responsible AI deployment.

Method

Implement structured AI use case assessments covering privacy, cyber, safety, and ethics. Establish an AI council for prioritization and risk management, and manage agent lifecycle for efficacy and model drift.

In practice

• Develop AI copilots for complex operational procedures.
• Optimize maintenance schedules using historical and real-time data.
• Standardize AI platforms for repeatable deployment patterns.

Topics

• Industrial AI
• Energy Sector AI
• Agentic AI Systems
• Data Governance
• Predictive Maintenance
• Operational Optimization

Best for: Investor, CTO, Executive, Director of AI/ML, VP of Engineering/Data, Consultant

Related on AIssential

• The AI Interview: Philippe Rambach, Schneider Electric — Scaling industrial AI requires a business-first strategy, rigorous validation, and fostering critical thinking.
• Getting to 82% Renewables: How AI secures our energy future — Scaling AI for energy transition requires addressing data fragmentation, IT/OT convergence, and trust through a platform-first, business-critical approach.
• How AI Is Powering the Next Industrial Revolution — AI can drive significant operational efficiencies and sustainability in resource-intensive industries by augmenting human productivity.
• Industrial AI for the Physical World: Siemens’s Peter Koerte — Industrial AI prioritizes precision and domain-specific data to optimize critical infrastructure, demanding high accuracy and value-driven data sharing.
• The AI Interview: Kaynaz Behdin — Stellantis drives AI for measurable business performance, integrating it deeply across operations with robust governance.
• FOD#139: Living Inside Kardashev’s Head: What Does It Mean For Us? — AI's increasing energy demands and agentic capabilities are transforming it into a critical, large-scale infrastructure.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by MIT Technology Review.