Rage against the machine?
As defined by the Oxford dictionary, ‘artificial’ is something “made or produced by human beings rather than occurring naturally, especially as a copy of something natural”. Intelligence’ is “the ability to acquire and apply knowledge and skills”.
An obvious antonymous phrase for these words used in conjunction is ‘unnatural stupidity’. The proponents and the potential beneficiaries, or indeed the victims, of ‘artificial intelligence’ (AI) seem content to accept the conflation of these terms without question.
Consciousness is a little more complex and its definition is necessarily more problematic than ‘artificial’ or ‘intelligence’. If consciousness relates to the fact of awareness by the mind of itself and the world, then responsiveness to one’s surroundings inevitably involves a sense of ‘knowing’ with the ‘other’ or within the self. In other words, the ability to understand and share the feelings of others; that is empathy.
The habit of saying things like ‘I need to recharge my batteries’ or ‘we are hard-wired for this or that’ indicates that we are already on the slippery slope towards believing programmable machines to be comparable and even superior to the human mind which is a dazzling self-organizing and conscious product of evolution. Some would say the mind is the creation of a divine being.
Bertrand Russell in “What is the Soul?” contends that physics “rendered matter too ghostly to be used as an adequate stick with which to beat the mind. Nevertheless modern science gives no indication whatever of the existence of the soul or mind as an entity; indeed the reasons for disbelieving in it are very much of the same kind as the reasons for disbelieving in matter.
Mind and matter were something like the lion and the unicorn fighting for the crown; the end of the battle is not the victory of one or the other, but the discovery that both are only heraldic inventions.” In the end, he says that while science diminishes our cosmic pretensions, it enormously increases our terrestrial comfort. “That is why, in spite of the horror of the theologians, science has on the whole been tolerated.”
Now, it seems we have, on the whole, abandoned faith in favour of supposed reason. Science therefore must serve competing ideologies and interests, including it may be presumed, those of artificial intelligence in as yet unimagined forms.
I accept that data isn’t information; information isn’t knowledge and knowledge isn’t wisdom. Computers, being unnatural and profoundly stupid, cannot understand this concept and therein lies the danger.
I’ve tried to establish here that that robotic machines are an artifact of human intelligence and that they lack consciousness, intelligence and empathy. Now, relying again on the Oxford dictionary, ‘intelligence’ can also mean “the collection of information of military or political value”. The late Stephen Hawking went so far as telling the BBC that “the development of full artificial intelligence, could spell the end of the human race”.
His concerns centred on the idea of machines imitating and even displacing [narrowly defined] human thinking. He worried that AI machines might “take off” on their own, modifying themselves and independently designing and building ever more capable systems. Humans, bound by the slow pace of biological evolution, would be tragically outwitted.
The advent and use of nuclear weapons soon gave rise to the peace movement with terms such as ‘doomsday machines’ entering popular usage.
As if to remind us of the potential for self-destructive madness that lurks within, chemical and biological weapons have since WW2 made their obscene appearance. The delivery of such horrors by means of drones and missiles guided by ‘artificial intelligence’ is no longer the realm of science fiction. The prospect of battalions of robot warriors marching relentlessly on ‘enemies of the state’, lacking in empathy and guided by ‘unnatural stupidity’ no longer seems fanciful.
However, one of the many things that separate us from machines is humour. A depressed employee tells her boss that she’s worried about being replaced by artificial intelligence. The boss kindly reassures her saying: “don’t worry, we already offered your job to a robot but he didn’t want it.”
I don’t find that very reassuring at all, do you?
Max Thomas, Dip. Agric. (retired) worked in the public sector and in private consulting on a range of land, water and waste management projects. He prepared guidelines for irrigation with recycled water for EPA Victoria and developed a number of Environmental Management Systems in the water industry.
Alan Stevenson
August 15, 2018 at 3:41 pm
Humour, even black relieves tension. Faith on the other hand can be more dangerous than any other concept because it does not rely on logic. As a Christian, I am asked to take as an article of faith that my body will be resurrected and I will live in an earthly paradise for ever. I know my body will rot, get eaten or burned; I also know the Earth is about half way through its cycle to destruction, yet every week the same mantra comes up. We are all able to hold conflicting concepts at the same time.
AI therefore cannot, by definition accept by faith alone. It will accept any data given provided this does not conflict with previous data – it is basically a logical device and can therefore be misled either deliberately or in error – garbage in, garbage out. In this sense, AI can make mistakes. I do not believe that any computer can be made aware of ethics or morality in the way we understand the concepts. Even Isaac Asimov showed that his three laws (later, four) were error-prone.
I heartily agree with Max in this regard. Too much reliance on AI can only led to our downfall. We must at all times have a human in overall control.
Max Thomas
August 8, 2019 at 9:10 am
Nice work Alan. I especially appreciated “we are all able to hold conflicting concepts at the same time.” It reminded me of a rule in good science that goes something like as follows: ‘An hypothesis that does not include reference to things which might refute it, is empty.’ It appears that my original piece on AI did contain the seeds of its own demise and might readily be refuted by intelligence. But I have to admit, I would prefer that it be human intelligence.