The convergence of research in the fields of AI, neurotechnology and soft robotics raises the possibility of creating something very similar to Andrew, the protagonist of Bicentennial Man written by Isaac Asimov in 1976, which blurs the differences between human and artificialbeings by Andrea Monti -Initially published in Italian by MIT Technology Review Italia
‘Robotics’ and ‘artificial intelligence’ are topics that, outside the fields of research and industry, are invariably treated with unbridled optimism or apocalyptic tones that, despite being meaningless, generate distorted perceptions at every level, including decision-making. However, the fact that in these areas the results are still at an embryonic level, does not detract from the fact that the publicly available proof of concept allows us to understand quite clearly where research and the market are heading: the construction of interfaces with the brain, with the skeletal and neuromuscular systems, creating not only prostheses made of ‘hard’ material but also using biological tissues that make the prostheses usable in certain sectors that involve direct interaction with human beings.
On these topics, MIT Technology Review Italiainterviewed Professor Shoji Takeuchi, from the Department of Mechano-Informatics, Graduate School of Information Science and Technology, University of Tokyo, expert in biohybrid robotics, microfabrication and tissue engineering. With his research team, Prof. Takeuchi is studying how biological components (for example, muscle tissue and neurons) can be integrated with artificial systems to develop new generation robots, biohybrid actuators and prostheses.
Professor Takeuchi, you and your group have just published a very interesting study in Science Robotics that demonstrates the possibility of using human muscle tissue to move biohybrid limbs. What inspired this research?
We were interested in the possibility of replicating natural movement thanks to the adaptability and energy efficiency of artificially reproduced biological muscle tissue. Conventional robotic actuators cannot completely replicate this function. So our objective is to bridge the gap between biology and robotics by integrating muscle tissue into more realistic and reactive robotic systems.
What are the main challenges in integrating multiple muscle tissues into a prosthetic limb?
I would say first of all maintaining tissue functionality, preventing necrosis in the muscle structures and being able to achieve sufficient contractile force. In our study we demonstrated that we can provide linear activation with high contractile force (~8 millinewtons) and high contractile length (~4 millimetres), which can be converted into flexion of multi-articulated fingers by means of a cable-driven mechanism. Another obstacle to overcome was certainly the coordination of multiple muscle actuators to achieve precise and synchronised movement.
Could you explain how biohybrid actuators differ from traditional robotic actuators in terms of robustness, efficiency and adaptability?
Biohybrid actuators are soft, flexible and self-repairing, but require a controlled environment to survive, whereas traditional robotic actuators are more resistant because they are mechanical. Another difference is that biological muscles generate energy through chemical processes of glucose breakdown and ATP utilisation, making them more energy efficient than mechanical equivalents that rely on external power sources.
How can we ensure the long-term functionality and durability of muscle tissue in artificial environments?
We use microfluidic nutrient delivery, biomaterial scaffolds and electrical stimulation to keep tissues active and functional. We are also investigating vascularisation techniques to mimic blood circulation and improve tissue longevity.
Is it possible to control and coordinate muscle contraction within these biohybrid systems?
Yes. Muscle contraction is controlled by electrical stimulation that mimics natural neuromuscular signals. In the future, we aim to develop a system that incorporates neuromuscular junctions to drive muscle activation directly through neural networks.
Do you think it is necessary to collaborate with neuroscientists to improve the neural integration of biohybrid limbs?
Again, the answer is yes. Collaboration with neuroscientists is critical to developing neural interfaces that enable direct brain or nerve control of biohybrid systems.
Is there progress in interfacing biological tissues with electronic or artificial intelligence-based control systems?
Recent advances in soft bioelectronics, optogenetics and neural signal decoding have improved the way we can interface biological tissues with electronic systems. These technologies could enable more precise and timely control of muscle contraction.
Does AI have a role in optimising movement and control in biohybrid limbs?
Certainly. AI can be used to analyse muscle signals, predict movement intentions and regulate control in real time.
How close are we to creating biohybrid limbs that equal or exceed the capabilities of natural ones?
The development of biohybrid limbs is still in its early stages and there are still obstacles to overcome in terms of resistance, strength and sensory feedback, which are still limited at the moment. However, in the future, research into tissue engineering, artificial intelligence-based control and neural interfaces should gradually bridge this gap.
In the long term, do you foresee biohybrid limbs surpassing synthetic robotic prostheses, or do you think both technologies will continue to coexist for different applications?
I believe they will coexist. Biohybrid limbs will perform better in terms of biocompatibility and naturalness of movement, while synthetic prostheses will probably remain superior in terms of durability, power and precision.
How do you think Japan’s leadership in biohybrid technology will influence the global robotics industry?
Japan has a strong foundation in robotics, microengineering and biotechnology, which places it at the forefront of research in these fields. The ability to integrate such technologies will enable Japan to lead innovation in next-generation prosthetics, high-sensitivity sensors and soft robotics.
Thinking about what Mori Masahiro wrote in his famous article Uncanny Valley about the difficulty of social acceptance of robots that are too similar to human beings, do you think that Japanese society is ready to accept biohybrid prostheses?
In Japan, we have a generally positive cultural perception of robots, which could facilitate widespread acceptance of biohybrid prostheses. However, the ethical issues and the psychological impact of these discoveries on people should not be underestimated. These will be the main factors to consider in order to understand how our society will welcome biohybrid technologies.