Industrial AI is moving from digital prediction to physical action. Foundation models, physical AI, and agentic systems now allow automation of complex tasks in industrial settings. Unlike software running in a server, these systems touch machinery where a wrong decision affects safety, reliability, and critical infrastructure.
The safety dilemma
Responsible deployment is now central to this shift. Arti Garg, chief technologist at AVEVA, asks how organisations can maintain safety and reliable operations while delivering on new capabilities. The challenge grows as newer AI systems become more capable yet harder to predict and explain.
Connecting data
Data forms the base for this transition. Industrial systems hold information across telemetry, service logs, and engineering documents. Newer technologies connect and correlate that information faster, giving operators real-time support when diagnosing problems. AI-powered robots can gather information in hazardous environments without requiring workers to enter them.
Garg notes that AI can help manage complex power systems as renewable generation grows. Organisations also need better ways to understand AI’s own environmental footprint. Garg works on an IEEE working group developing a standard methodology for measuring that impact across electricity, energy, resources, water, and carbon.
Governance and oversight
Greater autonomy requires new approaches to governance. AVEVA’s framework emphasises security, efficiency, and human safety and oversight. Garg argues that AI should augment rather than replace people in critical decision loops. Guardrails determine where automated systems can act and where human supervisors remain responsible.
The next phase brings industrial AI further into the physical world. This includes autonomous robots, drones, and AI-assisted coding that allows domain experts to build new applications. Realising that potential requires more than deploying new technology. Organisations will need to rethink business processes, establish appropriate safeguards, and give experienced workers new ways to apply their expertise.
“Autonomous systems, whether they’re robots or drones, are really going to change the way that we work in plants, in power systems, on mining sites,” Garg says. “In a way, that will make these types of operations more efficient, much safer for the human beings involved and more productive.”
What it means
Workers in high-risk industries will see fewer manual inspections in dangerous zones. Robots will gather data while humans monitor from safe distances. The shift demands strict guardrails to ensure machines do not override human judgment in critical moments. Organisations must update workflows to integrate these tools without losing accountability.
