Evolution: Beyond Darwin
Within this section. we shall discuss the implications of evolution-by-design, rather than natural selection. Subsequent sub-sections will present 3 stages of evolution connected with the hybrid AI paradigm:
Today, Darwin's ideas about natural selection have become the accepted principle behind evolution. Originally, the idea relied on the dating of fossil evidence, but now has the additional support of genetic science. However, if we turn the clock back before 1859, the year Darwin first published his ideas, there were two very different schools of thought, which we may label as `creationist` and `Lamarckian`. At the time, the creationist school represented mainstream Christianity and the belief that God created the Earth, sometime during 4004 BCE. Aspects of this idea still exist today, but mainly through links to more fundamentalist groups. On the other hand, the Lamarckian school represented an idea known as 'the inheritance of acquired traits' that nearly disappeared into history.
But which of these ideas might best reflect the future of human evolution?
Well, based on scriptures, the creationists might simply state that we must put our trust in their God while we wait for some apocalyptic `End of Days`. In contrast, Darwinists may simply point to natural selection and an outside chance that humanity might evolve into something better in a couple of million years or so.
But what of the Lamarckian school of thought?
Lamarck's theory of evolution was based on the idea that individuals could adapt during their own lifetime and transmit traits they acquired to their offspring. Offspring then adapted from where the parents left off, enabling evolution to advance. While this idea was to be superseded by natural selection, there is now the possibility that humanity could determine its own evolution. In this context, the meaning of `inheritance of acquired traits` might be translated into `evolution by design`. Of course, this idea will not be liked by the creationists and may well be opposed by the humanists. Clearly, the issue of humanity taking control of its own evolution may be perceived as a major departure from what some might call the `natural order of things`. Evidence suggests that it was the natural order of things that led modern man to migrate out of Africa some 140,000 years ago. This period of time, which accounts for some 7000 generations, saw human civilisation progress from the Stone Age to the Space Age. Surprisingly, for all this cultural evolution, there were no accepted major evolutionary developments in human physiology, including the brain.
So how does evolution account for today's world and in what form will it address the demands of tomorrow's world?
While natural selection may have been responsible for the evolution of human intelligence, it is clear that other processes have been at work. In 1976, Professor Dawkins outlined the concept of a `meme` as a unit of information. He suggested that memes could pass from one generation to the next in a process analogous to natural selection. While the spread of memes might be subject to a form of `survival of the fittest`, the rate at which memes spread is linked to attributes, which today could be described in the following technical terms:
- Data Accuracy
- Speed of Transmission
- Longevity of Storage
As such, the advent of language, writing and the printing press all represent major milestones in the ability of one generation to pass its knowledge to the next. However, civilisation had to wait for the advent of computers to see the biggest step function in this capability. Today, computers have evolved to a point where semi-automated information systems exchange terabytes of information around the globe everyday. This information now underpins all technology developments, which may in-turn open up the possibility for humanity to determine its own evolution.
In the wider context of the future evolution of humanity, it might be sensible to start with a very basic question:
What is the purpose of life?
From the perspective of natural selection there may only be one answer, i.e. survival, but is this goal sufficient to meet the aspirations of an intelligent sentient lifeform? Carl Jung (1875-1961) possibly captured the nature of this aspiration in the following quote:
"As far as we are able to understand, the only aim of human existence is to kindle a light in the darkness of mere being."
If mere being is not enough, then we must look for another answer. If the notion of an afterlife is rejected, then we must find purpose within life itself. Irrespective of how or why life has evolved from cellular survival towards intelligent sentience, we now have a better understanding and, occasionally, an appreciation of the wider universe.
As such, does the purpose we attribute to life also have to evolve?
If so, we may have to face up to a very disturbing possibility rooted in the fact that there is no evidence that natural selection was ever `programmed` to arrive or stop at humanity. If we accept this statement and choose to take on the responsibility for our own evolution, then we may also be forced to choose a path that may lead to our own extinction; at least in the sense that Homo Sapiens may eventually be superseded by `Homo Nexus`.
Along with the second law of thermodynamics, evolution suggests that change is inevitable and universal. While evolutionary change, in the form of natural selection, may have been so slow as to be unrecognised within a human lifetime, change was ever present. If so, humanity may only really have two options:
- be the victim of change or
- be the instigator of change.
Of course, many people who say they support change often do so with a caveat, i.e. it must not affect me. Unfortunately, circumstances surrounding our existence on planet Earth may be accelerating towards a point where painful change can no longer be deferred. While the increasing fragility of the global ecology is a major problem, a bigger problem may be technology itself.
Why is technology the problem?
In the next 100 years, humanity may have to face up to an ecological disaster on a scale that previously would have led to mass extinction. While technology has the potential to address some of these issues, it is unclear that humanity has the ability to understand the cause and effect brought about by the sum total of all its technologies. In the next 50 years, we could see today's information systems evolve to become virtually autonomous AI systems on which much of the world's population are dependent, but of which they have little understanding.
Artificial Intelligence (AI) is an acronym that has been around for over 50 years. How people react to AI depends on their understanding of two basic definitions:
- Weak AI claims that some thinking-like features can be added
to computers to make them more useful tools, but control is still
retained by humans.
- Strong AI claims that computers can be made to think on a level at least equal to humans.
As a broad generalisation, when people first hear these definitions they often consider weak AI to be just a matter of time and technology. While, in contrast, strong AI is often considered to be closer to science fiction than science fact. However, it is suggested that both types of AI have the potential to immediately affect the future evolution of humanity, both in terms of its physiology and its society.
But surely Weak AI is just another type of computer?
Yes, but as computer processing and the ubiquitous network become evermore pervasive, weak AI systems will expand into virtually all aspects of human lives, e.g. government control, industrial processes and home management. Eventually, there may be no such thing as a stand-alone device, as everything will be connected to, and monitored by, weak AI systems. As a consequence, a growing percentage of the population may become increasingly dependent on such systems in their everyday lives, to the extent that any new technology that helps them remain permanently in contact with these systems is not only accepted, but comes to be thought of as essential. As such, it may signal the first steps on a path, which sees the merging of humanity and technology into what might be considered as a departure from `homo sapien`.
OK, but will strong AI ever happen?
It is possible that strong AI may never be achieved. However, even if this is so, the probability of `Hybrid AI` should not be overlooked. Today, given our limited understanding of human consciousness, there are many sceptics who doubt the feasibility of ever being able to truly create sentient AI. However, hybrid AI would not require artificial sentience, as the essence of sentient intelligence could reside in a DNA-component evolving in-line with technology towards a genetic-AI hybrid. By this definition, hybrid AI is not science fiction, as the initial steps on this path have already been taken. Today, a brain-computer interface is the technology that allows neurons or sensory nerve tissue to be connected to silicon-based processors. Of course, it is recognised that this technology is currently in its infancy, but the potential for progress over the next 50 years is enormous.
The Genetic-AI Hybrid
Let us first get some idea of the possible timeframe in which a phased evolution towards an AI-hybrid could take place and why. On the Darwinian scale, any significant evolution takes place very slowly, i.e. millions of years. However, the potential for genetic science and AI technology to `evolve` could take place in a human lifetime. As such, the main factor governing hybrid evolution may be its acceptance by society at large. However, if the conditions to support human life become increasingly harsh, then the pressure for radical social change may also grow. Equally, there is the potential, over the next 100 years, for space exploration to expand into the solar system and require some people to live a major portion of their lives in an environment for which evolution never prepared them. Of course, there is every probability that military research around the world will be examining these possibilities, irrespective of the concerns or approval of mainstream society. If we accept even some of these speculative assumptions, then it is possible that some sections of society may start down an evolutionary path towards some form of genetic-AI hybrid.
An Evolutionary Path
The hybrid AI paradigm starts with Homo Sapien as the root of an evolutionary tree, which then branches out in different directions that one-day might lead to artificial life, not just AI. Although it will still be an evolution process of sorts, it may require less than 500 years for the following branches of this evolutionary tree to grow:
- Homo Computerus
- Homo Cyberneticus
- Homo Hybridus
The `scientific` names are fictitious and are only intended to convey the essence of the change being implied.
It could be argued that we have already taken this first evolutionary step. Although computers are not an integral part of our biological DNA, they are an integral part of our global society. Homo Computerus already has an information storage capacity that is billions of times greater than Homo Sapien, plus the ability to process and distribute this massive increase in information on a global scale. As more and more weak AI systems come on-line over the next 50 years, it will only accelerate the development of technology, but more importantly, the world population, estimated to grow to 8.9 billion by 2045, may now be totally dependent on its existence. As such, there may be no turning back without a total collapse of the global economy.
By this stage, remarkable enhancements to the human genome may have led to increased brain capacity. Even so, biological constraints may still mean that the brain may not be able to compete with the growing processing capability of technology. However, even though computer processing will evolve, over the next 50 years, to become self-learning within restricted problem spaces, it will not be sentient. Therefore, Homo Cyberneticus may be the logical step by which human sentience and creative imagination fuses with the ever-increasing processing capacity of technology.
The brain-computer interface would then allow near instant access to information and additional information processing. For example, if information from the optic and audio nerves could eventually be intercepted, and uploaded in real-time, any sights or sounds experienced by a person could be processed. The implications on our present-day perception of law and order would be profound. However, another by-product of the brain-computer interface could mean that a Homo Cyberneticus society could become effectively telepathic in the sense that it could directly share both sight and sound experiences over huge distance via the underlying AI network. This ability would represent a major crossroad in evolution, as Homo Cyberneticus would then be able to enter, and possibly immerse itself, in a world of augmented reality.
Up until this point, human appearance may not have changed in any appreciable way. However, Homo Hybridus would represent the start of a physical evolution to accompany the mental evolution within. Again, it may surprise some that this evolutionary process has already started. Today, in medical research laboratories around the world, people are already developing biomechanical systems in an attempt to alleviate disabilities resulting from limb amputation and/or paralysis. Today, these developments are still at an embryonic stage, as any bio-engineered limb is still awaiting the further development of a brain-computer-interface that would allow conscious control the artificial limb. In the future, having already accepted Homo Cyberneticus, society may have few qualms about accepting evermore-radical prosthetic enhancements to the human anatomy.