Automation helped shape the First Industrial Revolution – and it continues to evolve in today’s Fourth. While automation has long been part of the manufacturing landscape, recent advances in artificial intelligence, vision systems and robotics hardware are enabling a new generation of more intelligent and adaptable machines.
A new white paper from the World Economic Forum, Physical AI: Powering the New Age of Industrial Operations, explores how these developments are expanding the role of robotics - not just to boost efficiency, but to support greater flexibility and resilience on the factory floor.
Until recently, most industrial robots were designed for fixed, repetitive tasks in controlled settings. That’s beginning to change. With Physical AI, robots are gaining the ability to perceive, learn and respond to more complex environments while supporting a wider range of tasks and use cases.
This shift comes at a critical time. Manufacturers today are navigating a challenging environment shaped by rising costs, workforce shortages and shifting customer expectations.
But how did we get here? Understanding the evolution of industrial robotics offers important context for what’s next.
The evolution of industrial robotics
The application of Physical AI is the next step in a long evolution. We may think of robots as futuristic, but the earliest industrial robots go back to the 1960s. The term itself comes from the Czech word robota, which means forced labour or work.
The industrial robots of the past were rule-based – meaning they were explicitly programmed to execute repetitive tasks that require great precision and high speed, but lacked flexibility. Those systems have become a staple in sectors such as automotive and electronics, which have benefitted from the productivity increases robots brought to the shop floor.
For low-variability, high-volume tasks, these systems will continue to play a role, with their applications and abilities continuing to evolve.
We are now seeing the emergence of Physical AI through training-based robotics, where AI and machine learning are used to learn from simulated or real-world experiences. Unlike their predecessors, they don’t just rigidly follow a specific programme but can perform tasks that involve some level of variation, too. This makes these robots more adaptable for mid-volume and even non-repetitive production tasks. Crucially, their training can be virtualized, dramatically reducing deployment time and widening the scope of tasks that can be automated.
Context-based robotics mark the next evolution in intelligent automation. Like training-based robots, they are equipped with perception tools - from high-resolution cameras to tactile sensors - enabling them to “see” and interpret their environments in real time. What sets context-based robots apart is not how they perceive, but how they process and respond to unfamiliar tasks.
Key to these capabilities are powerful AI foundation models, which generate output from natural language prompts, integrating vision, language and action to understand their environment. They can take in the context they are operating in, ‘think’, make decisions autonomously and even plan. The white paper likens the extent of these skills to “human-level task intuition and planning”.
While these robots will still be far from the humanoid form factors we know from the movies, their appearance is also changing: quadrupeds, humanoids, mobile robots and many other shapes have emerged, extending the range of robotic applications.
That said, all three types of robotics – rules-based, training-based and context-based – will continue to be used in manufacturing. As part of diversified automation strategies, their deployment will be tailored to the needs of different production lines and types.
Why Physical AI and intelligent robotics are key for manufacturing
For manufacturers, robotic help could not come at a better time.
Supply chains remain fragile, exacerbated by geopolitical tensions, raw material shortages, and transportation bottlenecks. Market uncertainty further compounds these issues, threatening productivity, profit and resilience.
Rising raw material costs, energy prices and wages, along with workforce shortages and a growing skills gap, are equally adding to the sector’s challenges. Meanwhile, customer expectations demand more customization, faster delivery and sustainability.
Intelligent robotics links the digital and physical worlds to enhance operational flexibility to achieve this, but manufacturers must embed robotics in their long-term strategies, not just short-term gains.
Creating a workforce empowered to manage robotic automation
A skilled workforce will be essential to realizing this transition. According to the Forum’s Future of Jobs 2025 report, robotics and autonomous systems will be major sources of job displacement. But as the latest white paper on Physical AI suggests, this displacement is not just a disappearance – it is a transition. Along with AI and other digitalization, they will also drive the creation of new, skilled roles.
For example, machine operators will become robot technicians, logistics teams will coordinate mobile robots, maintenance teams will shift to predictive maintenance, and manufacturing engineers will focus on training and optimizing AI and robotics systems. An added benefit is that automating previously manual jobs will free people up to perform more meaningful tasks.
Successfully integrating intelligent robotics into the workflow requires a focus on workforce development and continuous learning. Reskilling and upskilling, as well as long-term workforce planning, will be essential to ensure that intelligent robotics delivers on its promise – not only for the business but also in social terms.
Physical AI in the real world
While intelligent robotics is still a young field, early adopters are already showing us the benefits of deploying the technology.
Amazon has over a million robots across its 300 fulfillment centres. They collaborate with human employees to handle repetitive tasks, like sorting, lifting and transporting packages. Robotic packing lines also help minimize packaging waste, supporting Amazon’s sustainability goals.
Orchestrating these systems has already led to impressive results in pilots, including faster delivery times and a 25% boost in efficiency. Managing the entire fleet of on-site mobile robots has increased travel efficiency by 10%. Amazon also created 30% more skilled jobs at its test site.
Meanwhile, Foxconn, an electronics contract manufacturer, is transitioning to what it describes as a "scalable AI-powered robotic workforce" to respond to rising labour costs and local manufacturing trends.
The company employs AI and digital twin technology to simulate and automate precise tasks such as screw tightening and cable insertion, which traditional rule-based robots previously found challenging.
Digital twin simulations have cut deployment times for new systems by 40%, while AI-powered robots have improved cycle times by 20-30% and lowered error rates by 25%. Operational expenses have decreased by 15%, and overall, AI-driven robots have demonstrated higher success rates than humans in complex assembly tasks.
How can manufacturers reap the benefits of Physical AI?
Physical AI is not the distant future. Intelligent robotics is already transforming manufacturing, and momentum will only increase. We will see more and more human-like capabilities emerge over time – even if the form factor isn’t always humanoid.
Manufacturers must act quickly to harness the potential the technology offers in response to the many challenges it faces, not least labour shortages, productivity and greater responsiveness to market and economic developments.
In contrast to using robotics in isolation, the World Economic Forum advocates for a layered automation strategy that integrates various types of robotics to achieve system-level intelligence.
While the pace of technological advancement is dazzling, companies must not get carried away. The focus should remain on people-first strategies to ensure sustainable, inclusive robotics integration. Manufacturers also need to leverage collaborative initiatives to share intelligence and confidently navigate this new era of automation.
Weforum