- Published on
Building a safer path to autonomous industrial AI
- Authors
- Name
- DeepHorus
Industrial AI is entering a more autonomous phase, says Arti Garg, chief technologist at AVEVA. Advances in foundation models, physical AI, and agentic AI are enabling automation of complex tasks across industrial environments. Unlike purely digital AI, these systems interact with physical operations, where unexpected decisions can affect safety, reliability, and critical infrastructure.
Data is one foundation for responsible transition. Industrial systems hold telemetry, service logs, engineering documents, and other disparate information. Newer technologies can connect and correlate it faster, giving operators real-time support when diagnosing problems, while AI-powered robots could gather data in hazardous environments without exposing workers.
Greater autonomy also requires new governance. AVEVA's responsible AI framework emphasizes security, efficiency, and human safety and oversight. Garg argues AI should augment rather than replace people in critical decision loops, with guardrails defining where automated systems act and where human supervisors remain responsible.
Sustainability is another part. AI can help manage complex power systems as renewable generation grows. Organizations also need better ways to understand AI's environmental footprint; Garg is involved in an IEEE working group developing a standard methodology to measure impact across electricity, energy, resources, water, and carbon.
The next phase could bring industrial AI further into the physical world. Autonomous robots and drones may inspect high-risk machinery, and AI-assisted coding could let domain experts build new applications. Realizing that potential will require rethinking business processes, establishing safeguards, and giving experienced workers new ways to apply expertise.
Garg says autonomous systems will change work in plants, power systems, and mining sites. She expects operations to become more efficient, safer for human beings, and more productive. The challenge is to pursue that autonomy while preserving safety, reliability, and accountability.
