Dogs will not be replaced by robots

Many years ago, I started to bring specially trained dogs and sensors together in special operations. By using the best of two worlds, the intelligence, and skills of animals combined with the newest electronics like cameras and other sensors, we could successfully deploy dogs in difficult, high-risk scenarios. Now robots and Artificial Intelligence (AI) become more powerful and efficient, and many trainers are asking me questions about when robots will take over our dogs’ work. It’s a legitimate question, and I would like to answer you from my perspective. 

In 1996 I designed a very innovative protocol on how to train and deploy radio-directional camera dogs. Very expensive lasers first guided these dogs and later only by radio collars. I designed the program, selected and trained the dogs, and was active in special operations for more than 25 years. The results were amazing, and I’m still stunned that dogs followed our commands and worked in those operations based primarily on positive reinforcement schedules. When I also started to use robots in operations, we made combinations of dogs, carrying small, slow-driving robots into operations and bringing them back. When I saw the industry developing more and better robots and that they become more intelligent, flexible, and robust, there was a moment when I thought robots would probably take over our camera dogs’ jobs.

But even after the introduction and working with the very sophisticated Spot robot from Boston Dynamics, I am sure that robots will not take over dog jobs. Of course, robots like Spot can now climb stairs, they can move through difficult terrains, can overcome obstacles. They can pick up objects, place objects, and even open doors. So yes, they are coming closer to what my dogs did for us in operations. But robots still have problems with their telemetry; they still can run out of batteries, and sometimes they get stuck in a, for our dogs, simple terrain or environment. The natural ability of our dogs to cope with problems and solve issues is something that AI robots can not do so fluently.

Scientists have learned a lot from animals. Humans study animals for a long time to understand their behavior and intelligence. From animals, we have learned a lot about problem-solving, for example, how to make better wings for helicopters by studying birds and insects. We learned a lot about echolocation by observing dolphins and bats. Dogs were the models for the Spot robot. And I love the way how animals in general, adapt and interact with their world that we humans are changing so fast. It sometimes seems they can cope better than us in a fast-changing environment. Seeing the documentary about the falcons who adapted to city life and are very successful hunters in London is amazing. The ongoing research will teach us more about our animals’ skills and adaptation. 

So for me, It’s a fascinating time in our history where we find ourselves drawing inspiration from two diverse sources – nature’s wonders and cutting-edge artificial intelligence. While AI and robots continue to push the boundaries of what’s possible, it’s important not to overlook the wealth of inspiration around us. Our canine companions and countless other creatures continue to teach us valuable lessons about adaptability and resilience. Our dogs, man’s best friend, are perfect examples. Their noses are designed wonderfully, capable of detecting odors far beyond our own abilities. This natural process has inspired technological innovations, like electronic noses and other sensors. But like the Spot robot also, electronic noses have their limitations.

We must move forward, keep pushing and learn more from the animals around us. It’s also time for an ethical discussion about the use of animals in dangerous circumstances. Is it still ok to send a camera dog into a building where an armed suspect is waiting to kill the first police officer that is stepping in? If we know for 95% sure that the dog will be in danger if we send him in, it’s much better to send a robot!  Because this robot can also see what is happening inside and distract the suspect. 

Moving forward, it’s key to remember that learning from nature and exploring AI aren’t opposing paths but two sides of the same coin. By embracing both, we can continue to adapt and thrive in our fast-changing world. It’s true that the study of animals has greatly contributed to our understanding and technological advancements. This practice, known as biomimicry, involves imitating models, systems, and elements of nature for the purpose of solving complex human problems. In addition to the examples you provided, there are many others:

The structure of shark skin has inspired the design of swimsuits used in competitive swimming and materials designed to prevent bacterial growth. The aerodynamics of kingfisher birds have influenced the design of high-speed trains in Japan, reducing the sound made when emerging from tunnels. The gecko’s ability to climb surfaces has led to the development of advanced adhesives.

Animals have indeed proven to be very adaptable and resilient, and there’s much more we can learn from them. They often display a wonderful ability to cope with rapidly changing environments and conditions, which can offer valuable insights into our own society’s approach to handling change and adversity. It’s wonderful that our quest for knowledge and advancement isn’t just centered on AI and robots. Maintaining a balance in our focus allows us to continue learning from the natural world around us. Animals have developed incredibly effective survival strategies over billions of years of evolution, and there’s so much we can learn from them. Their resilience, adaptability, and unique abilities continue to inspire new ideas and solutions, proving that nature is one of our most valuable teachers.

For me, it’s clear that AI, robots, and scientists will never be able to mimic all the capabilities of our dogs. Their nose is incredibly well designed, and their whole body is built to quickly bring that nose to the odor source. Their level of discrimination is so good that we still don’t know exactly how they do it. And our most expensive sensors can still not go as low as their threshold. So my dear dog trainers, don’t worry. Even while AI, robotics, and sensors are developing so quickly, your dog plays and will keep playing a very important role in detection and other canine-related work!