[Scientist’s Essay] Two Eyes on the Future in a Split Second

We are living in an era of mixed expectations and concerns about the future brought by artificial intelligence. It’s an era of marveling at the incredibly rapid pace of technological advancement and struggling to predict the future that AI technology will usher in.

Science and technology have a value-neutral aspect. Scientists pursue research out of intellectual curiosity and find joy in the outcomes that result from it. They have a deep understanding of technology and know its limitations. However, this understanding sometimes limits the freedom to imagine and speculate boldly about the impact of technological developments on future societies. On the other hand, artists’ imaginations are uninhibited by this limitation. They expand the boundaries of what scientists cannot fully depict, using the limitless power of artistic imagination to reignite scientists’ creativity. What might the future look like when neuromorphic semiconductor technology, which emulates the working principle of the human brain and implements AI in hardware, becomes a reality? I think it is possible to imagine together through the eyes of an artist and provide insights into the future through artwork.

The convergence of science and art might seem unfamiliar to scientists. However, when we reflect on history, there have been geniuses like Leonardo da Vinci, who simultaneously delved into science and art. Both pursuits may share a common goal. Both fields strive to unravel the world, analyze through curiosity-driven research, and present their stories with imagination and insight into their world using their language. Despite their different languages, understanding the world and envisioning the future might be similar processes.

Meanwhile, what does it mean to emulate the brain? Do we truly understand how the brain works well enough to mimic it? How can such a complex organism have intelligence? While a ‘spike’ is considered a key concept in brain information processing, are we perhaps being too simplistic? Can simplicity explain complexity? What paradigm might follow ‘spike’ next? Through the ongoing process of questioning fundamental inquiries and seeking answers, researchers repeatedly experience the wonder and mystique of human creation. If we replicate the brain in hardware, will we have truly comprehended humanity? Neuromorphic technology seems to be a discipline that necessitates exploring humans. It would be wonderful to experience a Eureka moment sparked by a brilliant idea that sparks like the ‘spike’ manifestation.

When new technologies emerge, they are often met with excitement and sometimes fear. While we recognize the benefits of technology, we are also cautiously concerned about its unpredictable side effects. However, I believe in the goodness of the people who develop and use the technology. Suppose neuromorphic semiconductor technology accurately simulates the brain. In that case, I will dare to imagine the potential for implementing AI technology that aligns with humane values. I hope that the AI technology of the future society, facilitated by neuromorphic semiconductor technology, will be a companion to a bright future for humanity.

Looking ahead a century from now, perhaps when people of an era in which neuromorphic semiconductor technology is so naturally used in artificial intelligence technology in daily life reflect upon the artwork of this era, they might have a sense of banality and be struck by it within their own time while admiring the artist’s insight into the future world presented through artwork from long ago. I confess that my accidental encounters with unfamiliar art have become precious memories in my research career, enriching my intellectual imagination.

Jongkil Park
Senior Researcher at the Center for Neuromorphic Engineering, Post-Silicon Semiconductor Institute, Korea Institute of Science and Technology
Jongkil Park received a Ph.D. in electrical and computer engineering from the University of California, San Diego in 2014 and is currently at the Korea Institute of Science and Technology. His research areas are neuromorphic semiconductor design that enables artificial intelligence by mimicking neural networks in the brain, and a spiking neural networks (SNN) algorithm for mimicking the efficient information processing mechanism of neurons and synapses in the human brain. The human brain is known for performing high-level cognitive functions while consuming less energy (20W) than a light bulb. A neuromorphic semiconductor can be used as the brain for intelligent mobile robots by designing low-power AI semiconductors that mimic the brain’s highly efficient cognitive functions.