Capgemini Research, the research arm of Capgemini, released the report Physical AI: Taking human-machine collaboration to the next level, Examine how advances in artificial intelligence are shaping robotics and the potential business value they may generate.
The report describes physical AI as the transition from traditional automation to systems capable of acting autonomously in real-world environments.
The opportunity has been widely recognized across industries, with strong interest from sectors such as high-tech, warehousing and logistics, and agriculture, and broad regional agreements among executives from the U.S., Europe and Asia-Pacific, the institute said.
Capgemini Research Institute stated that, supported by the convergence of technological advancement and market demand, physical artificial intelligence is moving from early experiments to tangible commercial results. The organization notes that the development of base models enables robots to operate more independently in complex environments, while simulation technology reduces the time required for training by allowing virtual learning at scale.
It further highlights the emergence of feedback loops where deployed systems generate data that improves performance over time. The institute said this progress is supported by edge computing, battery technology, falling hardware costs, new business methods such as robotics as a service, and connectivity improvements including private 5G networks and precise wireless positioning.
Capgemini Research reports that a majority of executives believe physical AI will enable applications that were previously impractical. Examples cited include hazardous operations, micro-logistics, pick-and-place tasks and on-site inspections, as well as more specialized uses such as adaptive assembly in manufacturing, support functions in healthcare and aged care and damage assessment in the insurance industry.
Capgemini Research says physical AI is also becoming relevant to reindustrialization strategies in Europe and the United States.
The organization reports that more executives see reshoring efforts as a factor driving interest in physical AI, with many organizations considering it a priority in their automation plans over the next three to five years.
The institute noted that mature robotic systems, including autonomous mobile robots, industrial robot arms and collaborative robots, are expected to see the strongest growth in the short term.
Labor constraints are cited as a key driver, particularly in sectors such as agriculture, retail, high-tech, logistics and automotive manufacturing. The report also suggests that physical AI may help improve operational flexibility to reconfigure production systems more quickly, while also improving safety and reducing physical stress on workers.
Capgemini Research reports that while many executives expect physical AI to be more widely deployed within the next five years, relatively few organizations have achieved large-scale implementation to date. The institute noted ongoing challenges, including limitations in technology and operational readiness, that continue to limit expansion efforts.
It added that despite interest in humanoid robots, their adoption is likely to continue for a longer period due to concerns about technology maturity, cost and training complexity, and uncertainty about return on investment.
The Capgemini Research report also points to social factors, noting that public acceptance may influence the speed of adoption. It highlights regional differences in sentiment, with European markets experiencing varying degrees of perceived resistance.
“Physical AI marks a shift from systems that describe the world to systems that can act in it. However, robotics has long been overpromised because early breakthroughs created expectations that the technology cannot yet meet,” said Pascal Brier, chief innovation officer and member of the Group Executive Committee at Capgemini.
“The difference today is not hype, but the convergence of AI, data and engineering maturity. As long as we focus on working at scale, the opportunity is real. Responsible, safe and progressive deployment of physical AI is critical to building trust, while design safety, transparency and human oversight are at the core of sustainable human-machine collaboration.”
