Complexity, Intelligence, and How to Ignore the Meaning of Life

Whether we enter a robot utopia, a robot smog or somewhere in between, we will close the gap between robotically aware and organically alive. Do we create new life forms, as Jansens suggests of his progeny?
This post was published on the now-closed HuffPost Contributor platform. Contributors control their own work and posted freely to our site. If you need to flag this entry as abusive, send us an email.

Click here to read an original op-ed from the TED speaker who inspired this post and watch the TEDTalk below.

Theo Jansen is a remarkable artist, engineer and visionary. His beach creatures, Strandbeests, pulsate rhythmically and elegantly, converting miniscule amounts of energy into undulating strides of synchronized, symphonic leg and knee swings. But Jansen's vision extends well beyond the creation of kinetic sculptural forms: he aims to unleash self-caring mechanical machines, living and moving in the intertidal zone, between the dune hills and ocean waves of the Dutch coastline. His focus on a narrow strip of sand where water meets land is especially appropriate because his philosophy explores a boundary zone: the space between complex machines and living organisms. That conceptual gap is shrinking as the breadth of robotic invention and more fine-grained scientific models of life begin to collide along their shared boundary. That gap was not always so narrow: twenty years ago, artificial intelligence operated at the exclusively cognitive level, centered on explicitly modeling human representation and reasoning, creating models of thought in computers that elaborately mimic those of humans. In ensuing decades, artificial intelligence and robotics have come to embrace the radical notion that artificial life might be deceptively simple - just maybe.

Nobel prize-winning scientist Herb Simon famously asked the question, what makes life seem complex? Is the animal inherently sophisticated, or is the animal's behavior just an elementary rhythm beating in a very complex echo chamber? Consider an ant on the beach, seeking to reach its food source. If you map its path along the sand, it would appear to be a circuitous route. Perhaps that apparent complexity is only the result of the ant's simple reactions to a complex beach environment. Simon's thesis: what if all animal behavior has this property? Perhaps our own apparent sophistication and intelligence has more to do with our environmental stimuli than with our disembodied selves. This idea curses the Artificial Intelligence inventor who is content to create a thinking brain in a box: if complexity stems from the contingencies of the environment, then true intelligence must be embedded- directly coupled into the physical world; so much for brains in a bottle.

Just as Jansen appreciates the power of elegant mechanism, so this desire for deep worldly coupling privileges mechanisms far above their earlier station in the eyes of human creators. -- Illah Nourbakhsh

Robotics caught onto this AI philosophy in the 1990's, with a pioneering proposal by roboticist Rod Brooks that intelligence may not even require a robot to have a model of the world in its memory banks. Just couple the robot to the real world so tightly that the real world is its model, with reactions that derive such sophistication from the environment itself that intelligence emerges as a side effect. Just as Jansen appreciates the power of elegant mechanism, so this desire for deep worldly coupling privileges mechanisms far above their earlier station in the eyes of human creators. Karl Sims helped trailblaze this path, demonstrating the beauty of mechanisms apparent in his awe-inspiring physical simulations of new robot swimming motions. By combining robotic joints in novel ways using evolutionary programming, he visualized remarkable new species that defy the existing precedents we witness in the natural world.

Today, Sims and Jansen are no longer lonely pioneers. Mechanical innovation has become far easier thanks to dynamic simulation, rapid prototyping and materials advances. We can make snakes that roll like wheels; robo-birds that fly like hummingbirds and dragonflies; walking robots that use zero energy to walk downhill; and hopping machines that run as fast as springboks. We may be on the verge of a "Cambrian explosion" in the evolutionary diversity of robotic brethren. A beautiful day may await, when a Technicolor future of stunning and elegant mechanisms herald the coming age of social robots. Or, as I also point out in Robot Futures, we may also be headed for Robot Smog: when everyone litters our one world with all the robotic creature they can create, threatening solitude with constant noise; civilization with chaos.

Whether we enter a robot utopia, a robot smog or somewhere in between, we will close the gap between robotically aware and organically alive. Do we create new life forms, as Jansens suggests of his progeny? Hans Moravec argues, similarly, that our robotic descendants fulfill humanity's evolutionary destiny, and that they will be superhuman, more alive than we ever can be. So, will robots be alive in our robot future? In Alone Together, scientist Sherry Turkle describes her daughter's reaction to a live Galapagos turtle at the Darwin museum exhibition: 'just use a robot turtle.' But, Turkle asked with surprise, isn't it important that this museum exhibit has a real, living turtle? Of course not, says her daughter, along with all the other children at the exhibit. Liveness is irrelevant, children seem to think; just use a robotic turtle that behaves like the old-fashioned kind. The young frequently see without adult bias, and in this case they are predicting a future of ever-increasing robot complexity taking on ever-greater numbers of skills and responsibilities that were once the exclusive province of people. Aliveness starts to lose its privileged position when machines can duplicate most of our behavior.

Jansen, Sims, Moravec and armies of do-it-yourself robot makers are striving to push the envelope of robotics from all angles: mechanical elegance, embedded sensing, massive computational power, Cloud-based distributed intelligences. Jansen's creatures are special; they stimulate our most basic animal recognition instincts through agility and dynamics, tapping our nerve center in a way that no computational Jeopardy-player or medical expert system ever can. His is a mechanical expression of beauty and intelligence, a tangible harbinger of our future mixed-species culture.

Ideas are not set in stone. When exposed to thoughtful people, they morph and adapt into their most potent form. TEDWeekends will highlight some of today's most intriguing ideas and allow them to develop in real time through your voice! Tweet #TEDWeekends to share your perspective or email tedweekends@huffingtonpost.com to learn about future weekend's ideas to contribute as a writer.

Popular in the Community

Close

What's Hot