Siemens and NVIDIA jointly promote industrial AI systems: From digital twins to autonomous factories, accelerating AI implementation in manufacturing

Siemens(Siemens) with over 175 years of industrial heritage officially partners with NVIDIA(NVIDIA) to deepen collaboration in the industrial AI field. The two sides not only integrate hardware and software but also further combine AI, simulation, digital twin, and automation comprehensively, aiming to create an industrial AI operating system that can “operate in real factories and be scaled for deployment.” Siemens CEO Roland Busch and NVIDIA CEO Jensen Huang jointly explain for the first time the implementation timeline, application scenarios, and tangible impacts on manufacturing, energy, and the global supply chain.

Industrial AI Deployment: From Decision Support to Action

Busch pointed out that the key change in industrial AI now is that new generation models not only provide suggestions but can directly represent human decision-making and execution, enabling systems to begin autonomous operation and self-adjustment.

He also mentioned that many customers have further developed manufacturing processes into digital twins, optimizing processes virtually before implementing them in real factories; AI is indeed already operating on production lines, just moving toward a higher level.

Scaling is the real challenge; barriers must be lowered for easier deployment and replication

Busch admitted that the real difficulty is not whether AI is feasible but whether it can “scale up and expand.” Key hurdles include:

• Whether customers have sufficient skills
• Whether systems are easy to deploy
• Whether rapid cross-factory and cross-industry replication is possible

Currently, industrial AI implementation still heavily relies on specialized personnel and complex integration. Therefore, Siemens is focusing on lowering the entry barriers to make deployment easier and usage more intuitive. He also emphasized that solutions are already being adopted in shipbuilding, heavy industry, and startups, indicating that market momentum is accelerating.

NVIDIA Accelerates Siemens Software, Integrating Design to Factory in a One-Stop Solution

Jensen Huang stated that this collaboration is not just symbolic but a deep integration across hardware, software, and processes. Key points include:

• Accelerating Siemens’ EDA software
• Accelerating physical and process simulation software
• Integrating AI, physical AI, and large models into Teamcenter and factory automation systems

This means that in future chip and system design, NVIDIA will rely more directly on Siemens’ simulation and digital twin tools; simultaneously, NVIDIA’s own factories and partners like Foxconn( (Hon Hai) can utilize this industrial AI operating system for production lines and factory management, creating a complete closed loop from R&D to manufacturing.

)Note: Teamcenter is Siemens’ product lifecycle management software, a digital platform connecting enterprise personnel, processes, and data. It integrates mechanical, electronic, and software design, BOM, and process management through a unified digital thread, helping companies collaborate across the entire product lifecycle from concept, design, to manufacturing and service, speeding up time-to-market and reducing development costs. EDA software uses computer-aided design (CAD) tools to automate complex integrated circuit (IC) and electronic system design processes, including logic design, circuit simulation, layout, and verification.(

Digital Twins Reduce Trial-and-Error Costs; Edge Inference Accelerates Efficiency

Regarding AI’s impact on the real world, Huang Huang used “Vera Rubin” as an example to illustrate that the complexity and cost pressures of systems have become so high that a new design approach is needed. This system integrates six chips, with a single GPU consuming up to 240,000 watts, achieving a tenfold improvement in energy efficiency and cost performance over previous generations.

His point is that if the entire system design and verification can be completed within Siemens’ digital twin, trial-and-error costs can be greatly reduced, turning “impossible” into “mass-producible,” and approaching a one-shot perfect solution.

Busch also added that AI’s battlefield is not only in data centers but also in whether low-latency inference can be deployed at the factory edge. Now, AI chips have entered controllers, industrial computers, and edge devices, enabling factories to make real-time adjustments and optimizations rather than post-analysis, further improving yield, energy consumption, and overall efficiency.

(Note: Edge devices refer to computers/controllers installed in factories, machinery, or on-site that can perceive, compute, and respond in real-time.)

Autonomous Factories and Energy Bottlenecks Coexist, Supply Chain Pressures Extend to Space

Both sides agree that the demand for autonomous and highly automated factories is rising, driven by labor shortages, yield improvements, better energy efficiency, and especially critical for the reshoring of U.S. manufacturing.

Huang Huang described modern factories as “giant robots,” where the biggest challenge used to be how difficult it was to teach robots and the heavy reliance on software personnel. The value of physical AI lies in making robots easier to “teach,” replacing extensive manual programming with demonstrations.

Regarding energy, Huang Huang straightforwardly stated that all industrial revolutions are limited by energy, and the AI revolution is no exception; therefore, each generation of products must be more energy-efficient. Busch shifted the perspective to the entire power supply chain, noting that the high demand for quality power in data centers has put pressure on generation, gas turbines, high-voltage transformers, and distribution equipment, potentially causing bottlenecks in some regions.

Extending the topic to the Chinese market, Huang Huang said that demand remains strong, mostly reflected indirectly through enterprise demand. Busch also mentioned that Siemens’ software investments will continue to expand, with acquisitions not ruled out.

Finally, they envisioned a longer-term future where space data centers could have energy and cooling advantages. If space-based manufacturing becomes feasible, the most suitable products would be intelligent computing power that can quickly transmit data back to Earth. Over the next 2 to 3 years, as AI, digital twins, and automation fully integrate, autonomous factories will no longer be just a concept but a new starting point for global manufacturing competition.

(NVIDIA Alpamayo Ecosystem Launch: Enabling AI Self-Driving Cars with Reasoning and Decision Explanation)

This article about Siemens and NVIDIA jointly promoting an industrial AI system: from digital twins to autonomous factories, accelerating manufacturing AI deployment, first appeared on Chain News ABMedia.