What do technological devices such as the radio and creatures that glow in the dark have in common? Short answer: Both the artifact and the organism have special features that have popped into existence more than once.
According to research on evolutionary biology, bioluminescence – the ability of living organisms to emit light – has multiple origins. This means that there isn’t a single common ancestor that first exhibited the trait and then passed it on to various descending lineages.
“Advanced technologies and biology have extremely different physical implementations, but they are far more alike in systems-level organization than is widely appreciated.”
Rather, different species, distantly related to each other in terms of their genetic ancestry, began flickering on their own. In this regard, it could be said that, evolution filed more than one patent for light-producing genes, as it appears in both fireflies and jellyfish among others. In both nature and technology, we observe something called “Multiples”, breakthroughs that occur repeatedly in absence of a single point of origin. Bioluminescence, for example, has emerged more than 40 times.
These new properties, incredible in their improbability of arising in the first place, are reproduced across the different branches of the tree of life. For this reason it is difficult, if not impossible, for researchers to pinpoint the first iteration of a given multiple. The identity of patient zero in the genesis of a new trait is often fiercely debated among scientists, not unlike people arguing whether it was Marconi or Popov who invented the radio.
While the Italian Marconi is widely credited as being the inventor of the groundbreaking apparatus, in Russia “Radio Day” is celebrated to commemorate Alexander Popov whom Russians regard as true originator of the radio. However both Marconi and Popov came up with the idea to use electromagnetic waves to transmit information independently. Similar to the fortuitous conditions that occasionally arise in nature, it seems that the beginning of the 20th century was pregnant with technological potential. It was this fertile climate that provided the optimal conditions for a machine that would enable a whole new method of mass communication.
Convergent Evolution in Biology and Technology
Just as the history of the 20th century can hardly be imagined without the impact of radio technology and the technologies derived from it, some fundamental structures in nature can also be labelled as archetypal. This is because the different developmental trajectories of an aimless evolutionary process appear to inevitably converge to form certain higher-level traits.
Despite all randomness, there are universal patterns in nature that stabilize out of the primordial chaos of living matter. These patterns – specifically Multiples – cross the domains of nature and technology. Ultimately, all human engineering is the reverse engineering of principles and mechanisms prefigured in nature.
As one of the pioneers working at the intersection of biology and the complexity sciences, John Doyle of Caltech, states: “Advanced technologies and biology have extremely different physical implementations, but they are far more alike in systems-level organization than is widely appreciated. Convergent evolution in both domains produces modular architectures that are composed of elaborate hierarchies of protocols and layers of feedback regulation, are driven by demand for robustness to uncertain environments, and use often imprecise components. This complexity may be largely hidden in idealized laboratory settings and in normal operation, becoming conspicuous only when contributing to rare cascading failures. These puzzling and paradoxical features are neither accidental nor artificial, but derive from a deep and necessary interplay between complexity and robustness, modularity, feedback, and fragility.” .
The novel art of biomimetics is humanity’s survival guide in a world teeming with mysterious life forms weirder than anything even the most gifted Sci-Fi author could come up with. Possibly the strangest thing about these results of nature’s tinkering is that they are alive.
Nature is truly an unimaginable enigma. The human cell, for example, reveals a dazzling complexity to the biologist looking at it under the microscope – its inner workings, arguably a far more interesting technology than anything we’ve ever invented. With everyone talking about the coming advent of AI, the collective attention has turned to nature’s most elusive creation: sentience.
The quest for AI sets humanity on a path towards understanding what it means to be a thinking being. In this sense, the AI labs around the world are the modern equivalent of the ancient temple of the Oracle of Delphi. Above the entrance was the inscription: “Γνῶθι σαυτόν (Know Thyself!).”
But the question still stands: why should the intelligence we are trying to build be artificial rather than natural? Without a proper understanding of the biological foundations of human cognition, the search for AI is going nowhere. If we are to learn anything from the existence of Multiples, we need to understand that the boundary between nature and technology is blurry and permeable.
So, we might as well aim for the real deal: Bio-Inspired Intelligence. Reconceiving AI as a techno-biological Multiple holds the key to turning soulless automatons into living systems that resemble actual human beings rather than ventriloquist puppets.
Technology: An Art and a Craft
Biomimetics is an approach whereby design solutions are gleaned from nature and mapped onto complex technological problems. The biomimeticist is someone who uses a natural phenomenon as a blueprint for the re-engineering of materials and processes. According to a recent issue of the Fortune 100 magazine, biomimetics is one of the most promising tech trends of 2017 .
So, having already outlined in broad brushes the philosophical vision behind a biomimetic approach towards AI, it is now time to turn to the nuts and bolts of constructing Bio-Inspired Intelligence.
The word “technology” comes from the Ancient Greek word “τέχνη” which has the double-meaning of “art” and “craft”. The art of biomimetics is encapsulated in a certain world view that looks at nature with awe and wonder. It chooses to shun the grandiose “theories of everything” proclaimed by the sages of the scientific establishment, who somehow manage to condense the history of the universe into a powerpoint presentation.
“Genetic information can be thought of as a repository of user manuals for wondrous inventions that make the jealously guarded industry secrets locked up in patent databases look like the scribblings of a child.”
The practitioner of biomimetics, an artist as well as a technician, looks at a miracle not as something to be explained away but as something to be replicated.
The craft of biomimetics relies on the utilization of advanced tools provided by cutting-edge science such as genomics and bioinformatics; tools that have only become available within the last couple of decades. Biomimetics can be thought of as 21st century alchemy, with the philosopher’s stone required to turn lead into gold having been there all along: inside of us, in our genome.
By decoding the ever-increasing number of genomes of different organisms, humanity has finally gained the literacy required to read the book of life. Genetic information can be thought of as a repository of user manuals for wondrous inventions that make the jealously guarded industry secrets locked up in patent databases look like the scribblings of a child. Nature’s patents are open-source and everyone feeling inclined to learn will have the chance to do so.
Deciphering Nature’s Patents
Starting around the year 2000 the high-throughput sequencing revolution fundamentally changed the way scientists engage problems in genome evolution, human health and computational biology. Today, next-generation sequencing tools allow the relatively cheap, fast and precise characterization of genes and noncoding DNA regions of various model (e.g. mice and flies) and non-model organisms (e.g. deep-sea creatures such as tube worms inhabiting hydrothermal vent ecosystems).
“Scaling of power and performance will change as computer designs transition to decentralized multi-core and distributed cyber-physical systems.”
The generated data, assembled from automated bioinformatic pipelines, can then be used to compare genomes of species across vast phylogenetic distances, from viruses and bacteria to unicellular eukaryotes and multicellular organisms.
In fact, comparative genomics was initially used to deﬁne the minimal gene set of the Last Universal Common Ancestor (LUCA) of all life. It allowed scientists to constrain and, on theoretical grounds initially, reverse engineer an ancient minimally complex cell by producing a list of essential transcriptional and translational cellular parts.
From the origin of life to the emergence of multicellularity and complex nervous systems, comparative genomics has yielded unique insights into the evolutionary mechanisms underlying large-scale evolutionary events. As such, it presents a powerful tool in understanding major transitions in biological complexity.
Towards Neuromorphic Computing
In order to initiate an evolutionary “major synthetic transition” in biomimetic computation we have to comprehend in more detail how nervous systems initially evolved and developed. By using comparative genomics we can begin to reverse engineer the first fast synapse that originated more than 600 million years ago and subsequently complexified during the Cambrian Explosion, an enigmatic large-scale speciation event during which all of today’s animal complexity emerged.
The lessons learned can then be applied to the design process of our own technological implementations, which require a deeper physiological understanding of the neural substrate of learning and chemical computation within the brain.
The evolution of both biological and technological synapses (in neuromorphic systems) appears to be constrained by a single energy-time minimization principle that governs the design of many complex systems that process energy, materials and information.
Research by Melanie Moses (University of New Mexico) on scaling relationships in biological and technological systems once again stresses the profound similarities to be discovered between the two domains: “Just as the evolutionary transitions from unicellular to multicellular animals in biology are associated with shifts in metabolic scaling, our model suggests that the scaling of power and performance will change as computer designs transition to decentralized multi-core and distributed cyber-physical systems.” 
The question we are facing with our century’s computational architecture is whether we are actually staying true to nature’s lead or creating a clumsy “artificial” intelligence that is not based on how computation is performed in animal brains.
Outlook: Fusing Biomimetics and Comparative Genomics
Focusing on comparative genomics, the next installment of this series will outline how ongoing scientific work in various fields is revising our understanding of early nervous system and synapse evolution in animals and, at the same time, promises to give insights into truly, i.e. evolutionarily-based, biomimetic neuromorphic computing architectures.
At the same time, our method may facilitate the discovery of universal principles underlying complexifying biological and technological systems.
1. Csete, M.E. and J.C. Doyle, “Reverse engineering of biological complexity.” Science, 2002. 295(5560): p. 1664-9.
3. Moses, M., et al.: Energy and time determine scaling in biological and computer designs. Philos. Trans. R. Soc. Lond. B Biol. Sci. 371 (1701), 20150446 (2016).
There are several instances of developers borrowing ideas from nature to create new technologies. A pretty cute example is that color displays for e-readers were inspired by butterfly wings. Researchers from Qualcomm MEMS Technologies came up with the first full-color, video-friendly e-reader prototype, after studying the way butterfly wings glimmer in the bright light. It works by reflecting light instead of transmitting it; this allows the battery life to last longer and makes it easier to read the screen in bright sunlight.
To contact the editor responsible for this story:
Margarita Khartanovich at [email protected]
- Artificial Intelligence Isn’t Ready to Take Over From Doctors and Nurses, Just Yet
- Machine Learning Vs. Analysts: Will AI Eventually Replace Data Scientists?
- Why The Danger of Deepfakes Is No Danger At All
- Sarcastic Robots? How Deep Convolutional Neural Networks Are Making AI Worryingly Human
- Traffic Lights in The Sky: Flying Cars Can Appear Sooner Than You Think