On the PAI Methodology
"The PAI methodology inherently combines the considerations of materials, design, and manufacturing. Robots developed using PAI may autonomously perform tasks and maintain homeostasis in unstructured environments by exploiting the physical and computational features of their bodies along with the computational abilities of their "brains". Akin to biological organisms, PAI robotics can either replace digital AI or synergistically complement it by interfacing to a "brain". Many small robots, or those with limited computational functionality, would not include a dedicated central "brain," and their body's computation will guide their performance. Similar to the principles of natural diversity, PAI synthesis refers to robotic systems of any functionality, shape, size, and habitat, with particular emphasis on integrating chemical, biological, and material based functionalities."
On the PAI Methodology
"The PAI methodology inherently combines the considerations of materials, design, and manufacturing. Robots developed using PAI may autonomously perform tasks and maintain homeostasis in unstructured environments by exploiting the physical and computational features of their bodies along with the computational abilities of their "brains". Akin to biological organisms, PAI robotics can either replace digital AI or synergistically complement it by interfacing to a "brain". Many small robots, or those with limited computational functionality, would not include a dedicated central "brain," and their body's computation will guide their performance. Similar to the principles of natural diversity, PAI synthesis refers to robotic systems of any functionality, shape, size, and habitat, with particular emphasis on integrating chemical, biological, and material based functionalities."
What is Physical AI?
Physical Artificial Intelligence (PAI) refers to the theory and practice of creating physical systems capable of performing tasks typically associated with intelligent organisms.
Nature Machine Intelligence, 2(11) 2020 The full text of the paper is available here
The Origins of Physical AI
In 2020, Aslan Miriyev and Mirko Kovač introduced and defined Physical Artificial Intelligence (PAI) in their Comment Skills for Physical Artificial Intelligence, published in Nature Machine Intelligence. [PDF]
They presented PAI as a scientific framework for creating intelligent physical systems through the co-evolution of body, control, morphology, actuation and sensing. The methodology integrates materials, design and manufacturing to create autonomous robots capable of functioning in complex, unstructured environments.
The authors identified five foundational disciplines of Physical AI: materials science, mechanical engineering, computer science, chemistry and biology. Recognizing that creating such machines demands expertise across conventional disciplinary boundaries, they also proposed a methodology for developing the multidisciplinary skills needed by future generations of researchers.
The November 2020 issue of Nature Machine Intelligence featured their work on its cover, under the title Crafting artificial intelligence. The journal also published a dedicated Editorial, Materializing artificial intelligence, examining the scientific background and future promise of Physical AI.
Cover of Nature Machine Intelligence, November 2020.
Based on the "Hands of creation" image, showing human hands creating PAI by using skills from 5 disciplines. Image credit: Aslan Miriyev, Mirko Kovac / Empa and Imperial College London; Cover design: Karen Moore.
Physical Artificial Intelligence (PAI) is a multidisciplinary field at the intersection of five main disciplines: materials science, mechanical engineering, computer science, biology and chemistry. Adopted from "Skills for Physical Artificial Intelligence", Aslan Miriyev, Mirko Kovac, Nature Machine Intelligence, 2, 658-660, 2020.
Physical Artificial Intelligence (PAI) refers to the theory and practice of synthesizing lifelike robots. This is achieved by using skills from the PAI's five core disciplines to evolve a fully-functional robot (in the illustration: robotic bee). The process incorporates the considerations of materials, structure, sensing and actuation, and results in a fully autonomous multi-functional robot. Adopted from "Skills for Physical Artificial Intelligence", Aslan Miriyev, Mirko Kovac, Nature Machine Intelligence, 2, 658-660, 2020.
From a Scientific Concept to a Global Technological Direction
Six years later, Physical AI has become one of the defining visions for the future of robotics and intelligent machines.
What began as an interdisciplinary scientific framework has grown into a global research and industrial agenda. Major technology companies, robotics manufacturers and research institutions are advancing Physical AI through different approaches, from AI models capable of reasoning and acting in the physical world to the development of new robotic bodies, intelligent materials and integrated autonomous systems.
NVIDIA is building an extensive Physical AI ecosystem involving leading robotics manufacturers and developers. Google DeepMind is advancing robotic intelligence through its Gemini Robotics models. The World Economic Forum has examined Physical AI as an emerging force in industrial operations, while MIT hosted a dedicated international Physical AI Summit in October 2026.
These developments illustrate the breadth of the field. Some approaches emphasize computational intelligence, learning and autonomous decision-making. Others advance the physical technologies through which intelligence can be realized. They represent different parts of the broader Physical AI vision, although not all embody the complete integration of physical and computational intelligence proposed in 2020.
Physical AI is now influencing how researchers and industry envision the next generations of robots, industrial automation and autonomous machines. Its growing prominence reflects a fundamental shift in technological ambition: from developing AI systems that primarily process information to creating intelligent physical systems capable of acting in the real world.
Physical AI at PAI Lab
Founded in 2021 by Dr. Aslan Miriyev, co-originator of the Physical AI concept, the PAI Lab advances Physical AI through the development of materials, manufacturing technologies, functional systems and autonomous robots, with the long-term goal of creating sustainable human–robot–nature ecosystems.