What will it take to scale Industrial AI?
Scaling Industrial AI is difficult because in the physical world, "mostly right" isn't good enough. A mistake on a production line or power grid can interrupt output, damage equipment, or create safety risks. Industrial AI has to understand industrial context, standards, and constraints, and deliver on safety, uptime, and quality.
That's especially challenging in existing industrial facilities, where machines and systems from different eras and vendors were never designed to communicate with one another.
One answer is to design and test changes virtually before implementing them in the physical world. Digital twins allow companies to model products, processes, and entire facilities, simulate how changes will perform and optimize them before putting them into operation.
That approach is already producing tangible results. At one U.S. Gatorade plant, Siemens' digital twin helped PepsiCo improve throughput by 20 percent within 90 days while virtual designs unlocked double-digit capital-investment savings.
Scaling further will require interoperable technology and trusted data that businesses of every size can adopt, particularly the small and midsized manufacturers at the heart of the American industrial base.
But putting AI to work across the physical economy requires more than technology. It requires the infrastructure to support it—and, above all, reliable power.
Can America's grid power AI and U.S. manufacturing?
Yes, but we need to move faster.
AI, advanced manufacturing, electrification, and onshoring are driving electricity demand higher after decades of relatively flat growth. Meeting that demand will require abundant, reliable power and a more modern, flexible grid.
But this isn't simply a question of generating more electricity. Earlier this year, I spent time with mayors, utility partners, and data center developers discussing the challenges facing communities on the front lines of AI infrastructure investment. The questions are practical: What will this mean for electricity bills? How will new facilities affect the surrounding community? And how can we accommodate growth without putting additional strain on existing infrastructure?
We cannot solve the challenge solely by building more physical infrastructure. Grid software, distributed energy resources, and demand flexibility can help utilities improve reliability and get more capacity from existing assets. More efficient cooling technologies and intelligent energy management can also help data centers operate more efficiently and reduce their impact on surrounding communities.
That same thinking applies to factories and industrial facilities across America. Reindustrialization isn't only about building new plants. It's also about making better use of the infrastructure and industrial assets we already have.
Meeting these challenges will require technological innovation and collaboration among industry, utilities, communities, and government. New generation and transmission projects still must navigate overlapping federal, state, and local permitting and interconnection processes. Greater coordination, faster decisions, and more certainty can help bring critical power infrastructure online at the pace America's economy requires.
How should we prepare the skilled workforce for AI?
Before we even consider how AI might change manufacturing jobs, we need to address the hundreds of thousands of positions already unfilled. At Siemens, we recognize that workforce development isn't just about preparing our own employees. It's about supporting the industries we serve and the communities where we operate.
That requires multiple pathways into industrial careers—not just traditional four-year degrees, but apprenticeships, community college partnerships, and industry-recognized credentials. Siemens has committed to training 200,000 electricians and manufacturing experts by 2030. Our Expedite for Vets program helps military veterans translate their experience into skills for industrial careers.
The Siemens Foundation is also helping expand training capacity. Through Careers Electric, the Foundation is bringing together employers, educators, and workforce organizations to expand training opportunities for the electrical industry, including through partnerships with Wake Tech and 10 additional North Carolina community colleges.
At our Fort Worth facility, Siemens brought in retired teachers to help create a training program for people without previous industrial experience. Workers from hospitality, customer service, and agriculture have learned skills for today's factory. It's an example of how we can open doors to industrial careers for people who might never have considered them.
What would Industrial AI success mean for American growth?
Five years from now, the clearest measure of Industrial AI success won't be how much we talk about AI. It will be how broadly we deploy it across the economy.
The real test is whether a small manufacturer operating in a 40-year-old plant can deploy Industrial AI, power it reliably, and find people with the skills to use it. When that becomes routine, we'll know we're turning American innovation into industrial growth.
Ultimately, that's what matters: translating technological innovation and investment into greater productivity, stronger American industries, and more opportunities for people and communities across the country. The path to super intelligence runs through the AI we are deploying throughout industry today.
Published: October 27, 2026
