Generative Bionics presents Gene.01: the Italian humanoid robot that 'feels' touch before contact
Generative Bionics has announced Gene.01, a humanoid robot developed for industrial applications that aims to combine physical artificial intelligence (Physical AI), advanced sensor technology, and an open-source approach to software development.
The Italian company claims to have completed the project in just six months, moving from the initial concept to a fully operational platform designed to work closely with human operators. One of the distinctive elements of Gene.01 is the presence of a multisensory skin distributed throughout the robot's body. The system can detect touch, proximity, force, and temperature, allowing the humanoid to perceive its surroundings and anticipate the presence of people even before physical contact. According to Generative Bionics, this architecture enhances safety during human-machine collaboration and makes interaction in production contexts more natural.
The ability to measure force is also a key element for the robot's training. In addition to replicating observed movements, Gene.01 can learn the intensity with which to perform certain operations, an information that cannot be solely derived from video demonstrations. This approach is intended for activities where the precision of applied force is crucial, such as object manipulation or specialized industrial operations.
At the project's core is a proprietary Physical AI platform defined as "physics-native," in which mechanical design, motor control, and artificial intelligence algorithms are jointly developed from the early stages. The goal is to optimize both the body of the robot and its movement capabilities using measurable parameters, such as ergonomics and walking stability, rather than separately intervening on hardware and software.
Alongside the physical robot, Generative Bionics has decided to publish the digital twin of Gene.01 in open-source mode. The company claims that this is the first humanoid robot model distributed directly into the major software ecosystems used by Physical AI developers. This will allow for simulating the robot's behavior, developing new applications, and testing algorithms before the hardware becomes available, thus accelerating development processes.
Gene.01 is not presented as a single product, but as a platform from which different humanoid robots tailored to specific industrial sectors can be derived. For each application, the company plans to customize three main aspects: the Physical AI model, the external design, and terminal elements such as hands and feet. This setup is expected to enable the creation of a common hardware base adaptable to contexts such as logistics, manufacturing, security, inspections, and healthcare.
The first concrete use of the platform involves collaboration with Fincantieri, announced a few months ago. The two companies are working on a version of Gene.01 aimed at welding activities in shipyards, an environment characterized by complex operating conditions and high safety requirements.
On the industrial front, Generative Bionics also emphasizes its willingness to build a European ecosystem for Physical AI, leveraging local production chains, regulatory compliance, and components developed on the continent. This vision includes a partnership with the German company Synapticon, aimed at developing a European implementation stack that includes customized actuators, assembly, calibration, and safety testing.
The company also highlights that nearly half of its approximately one hundred employees hold a PhD and come from experiences gained on projects like iCub, iRonCub, and ergoCub at the Italian Institute of Technology. Generative Bionics has raised a total of 70 million euros in a seed round led by CDP Venture Capital, with participation from AMD Ventures, Duferco, Eni Next, RoboIT, and Tether, funds that will be aimed at developing the next generation of Physical AI platforms for industrial applications. Gene.01 will make its public debut in the United States during AMD Advancing AI 2026, scheduled for July 22 and 23.