Schneider Electric, Nvidia, HPE launch Industrial AI Agents at the Edge

The collaboration stitches together three different kinds of expertise: Schneider Electric's industrial automation and control systems, Nvidia's AI computing, and HPE's edge infrastructure. That combination is what lets the agents sit inside a plant and work with equipment and control systems that are already there, rather than arriving as a robot. The companies frame this as "physical AI" for industries such as energy, chemicals, pulp and paper.

The starting scope is deliberately narrow — monitoring OT, cybersecurity and energy consumption — with each agent type drawing on different data. OT agents look at equipment configurations, alarms, maintenance histories and real-time operational data; cybersecurity agents scan network activity and operational events for abnormal behavior; energy agents examine production and energy data for savings.

Schneider Electric points to CMPC's three years of experience with its OT Monitoring System as evidence the underlying monitoring works. Matías Jory of CMPC describes early issue identification as critical to protecting production, and sees potential for AI agents to help specialists investigate faster while keeping people in control of critical decisions. The longer-term aim the companies describe is for industrial AI systems to continuously monitor operations, identify problems and recommend or make adjustments — then check whether those actions produced the intended result.