Physical AI: The Next Great Manufacturing Shift

Manufacturing is at an inflection point. The industry’s workforce is ageing, and it has not yet cracked the code on attracting the next generation of workers.

At the same time, AI tools present a host of opportunities, including those that may help address the workforce challenge.

Article continues below

Michael Weller

Practice Leader for Verizon Business’ Manufacturing, Energy and Utilities (MEU) team.

In my conversations with manufacturing leaders, I hear the same tension repeatedly: they recognize the potential, but they’re wary of the risks.

The manufacturers that will pull ahead in 2026 are those that jettison their assumptions and lean into the next-generation tools that have the power to optimize the factory floor – and attract the young workers needed to run it.

collaboration between humans and autonomous machines. Computer vision is transforming quality control by providing a real-time view of factory assets, anticipating and preventing collisions, spills, and other errors before they occur.

Combined with digital twins and sensors, this technology can identify and predict machine faults, enabling plant managers and engineers to intervene before minor issues escalate into breakdowns. The result: extended machine lifetimes, reduced downtime, and improved operational efficiency.

Of course, many manufacturers remain uncertain – not opposed to AI, but unsure where to direct their computing power for maximum impact. The advent of Industry 4.0 and IoT devices, which paved the way for Physical AI, also created new potential for cybersecurity risks, reinforcing that caution.

But this is where I see the most productive shift in mindset happening right now. Manufacturers are increasingly recognizing that AI can be the bulwark against cyberattacks, not the source of risk.

A good example of this is what is sometimes referred to as “AI shells” – AI layers that wrap around legacy systems, infer the types of security risks those devices and systems have been exposed to, and act as a protective barrier to prevent compromises.

This is especially critical given that many legacy manufacturing systems cannot easily be patched. While digital innovation can sometimes create new vectors for threat actors to exploit, AI can also provide levels of protection that were simply not possible before.

After years of cautious experimentation, innovations like these are finally moving “out of the drawer” and into real-world deployment.

visualization are immediate. Engineers can peer inside machines using digital overlays to identify faults, rather than having to physically open them up. But the workforce implications may be even more significant.

Enhanced visual aids – from 3D schematics to just-in-time training – are powerful educational tools. Younger workers tend to be visual learners, and these environments speak their language in a way that traditional manufacturing settings simply do not.

asset management, and genuine operational flexibility.

What makes this moment different is the role of automation software. Most automation today remains hardware-bound, meaning upgrades require replacing physical infrastructure – a prohibitively expensive proposition for manufacturers who have invested heavily in existing systems.

Software-defined automation changes this equation significantly, allowing manufacturers to modernize through updates and upgrades without extensive overhauls. When combined with wireless connectivity, it creates an environment where connected workers, visual technologies, and mobile equipment can operate together seamlessly.

The environmental case is equally compelling. Wireless technology eliminates copper cabling and reduces network power consumption. Through our work with manufacturers, we’ve seen that a single cellular antenna can typically displace between three and ten Wi-Fi access points – a reduction in cabling and energy consumption that, across large facilities, can add up significantly.

The Convergence Is Here

This is not just a shift from physical to digital. It is an evolution from rigid systems to adaptable ones. Many of these technologies have already been introduced in manufacturing individually, but what previously existed in pilot programs or siloed use cases is now coalescing into something far more impactful.

What I find most compelling is the synergy: visual-first environments, intelligent infrastructure like Physical AI, and wireless technologies are not just complementary – they are mutually reinforcing.

Manufacturers who embrace this convergence can significantly improve productivity and safety, and build the workplaces that the next generations actually want to join. That is perhaps the most consequential shift of all.