This is an excerpt from Philosophy as Dinner Party Chatter, a forth coming publication by the Parmenideum
Given that both humans and the Net rely on energy for their survival, if the Net ever became “aware” of a future energy crisis, would we expect it to do the honourable thing and switch itself off?
Few of us, at the moment, have reason to ask whether or not the Net is conscious. It’s just a dumb computer network, albeit a very big dumb computer network. Besides it’s made from wires and silicon chips and runs on electricity. How could such an entity ever become conscious?
Not so fast. For some people the key words here are precisely: big, computer, network and electricity, which according to the Computational Theory of Mind are just the kind of ingredients you’ll need if you want to cook up a thinking, perhaps feeling, self-aware entity that isn’t necessarily made from soft carbon based tissue (or “wet-ware” as the AI expert Marvin Minskey calls it).
The computational theory more or less states that any substance with a sufficient number of working parts, arranged in the right way and with sufficient complexity, can produce consciousness. Consciousness will arise from the parts and connections as they perform calculations, almost as a by-product. You could use tin cans or doors connected by strings as a method of performing computation and it wouldn’t matter, as long as their configurations and operations were complex enough and set up the right way.
You could also use chemically fired neurons, which, as it happens, are precisely what humans possess in the form of a couple of pounds of grey and white matter safely housed between our ears and behind our eyes.
The hundreds of billions of neurons in the brain are connected through trillions of synapses and, as the theory goes, because there are so many of them and because they communicate through such a huge network of fast electro-chemical connections, they are not only able to perform stupendous feats of computation, they actually generate the phenomenon of consciousness – though perhaps albeit as an accidental (epiphenomenal) by product.
Some computationalists such as cognitive scientist and philosopher Daniel Dennett go even further than that, and claim that once a critical complexity is reached in any ordered organisation of matter, some form of consciousness is inevitable. As an interesting example, if this hypothesis is correct many animals with complex nervous systems probably possess some form of consciousness, and artificial systems that possess sufficient organisational complexity would also have some level of consciousness.
The big question though, is whether other complex non-biological entities are capable of the type of consciousness that we as “wet ware” experience. That is, the type of consciousness that experiences colour, or pleasure, pain, hope, saltiness, sweetness, hardness, love and so on. This type of consciousness is often referred to as qualia consciousness, or “phenomenal consciousness”, terms for expressing the apparently ineffable feature of – as philosopher Thomas Nagel puts it – knowing what it is like to experience phenomena such as colour, saltiness, sweetness, pain or hope for example. That is, knowing what it actually feels like to eat bacon or honey, touch a hot cinder, or wait for results of a medical test.
For such experience you do need to have an appropriate sensory system, but also an appropriately complex neurological system for abstract processing. Both together get you to the “what it is like” high-end of the consciousness spectrum. Many scientists in the field of AI do not view the simulation of either sensory or abstract processing as a problem. It is, they maintain, only a question of sufficient complexity.
Computers can in fact already be made aware of their surroundings with appropriate sensors, and can be programmed to respond in certain ways to those surroundings Yet, so far, no one has succeeded in creating an artificial system complex enough to generate self-aware consciousness, and no one has explained how consciousness actually arises from pure computation, what it consists of, and where it resides. What are thoughts for example? So it is worth bearing in mind that the theory is not only still just a theory, it so far also fails to explain what consciousness actually is.
However, the alternative to any physical explanation to consciousness is just one: dualism. If a physical system such as the brain is not the source of its own consciousness then the only other explanation is that something else is responsible for it. That “something” has traditionally been a “something” that isn’t “something” and is in fact an immaterial entity usually referred to as the soul. This soul enters a material body and makes it conscious. This idea was first put about by the Ancients in both Eastern Mysticism (Upanishads) and Greek philosophy (Socrates), and later given serious consideration by French Philosopher Rene Descartes in the 16th Century.
More recently other dualistic explanations have arisen. One notable theory is property dualism, which asserts that though nature consists of just one substance – matter – this matter can display both physical and mental properties. “Naturalistic dualism” is another notable theory proposed by philosopher David Chalmers, in which he posits the existence of a separate material substance with purely mental properties yet which is still part of the “existing physical ontology”.
But the dualist stance may have more problems with it than the computational, not least the fact that dualism can likely never be tested with anything physical. Quite obviously, there are no means through which either purely mental or purely immaterial (Cartesian) entities can be tested by the physical methods of something even as reliable as our current science. In the case of Cartesian dualism, the only method that can deal with it is faith, and that is no proof at all, as even the faithful admit. As for property and naturalistic dualisms, some new methodology is required, one that expands the current physical methodology into dealing with the hypothesised new “ontology of the mental”.
And of course one other major problem with Cartesian dualism is how on earth does an immaterial entity interact with a material object such as the body-brain assembly?
This is where computational theory finds its ground. It is not simply an explanatory theory that requires some way of validating it. Whilst like its rivals it offers an explanatory hypothesis, unlike its rivals it does not rely on purely observational means to validate itself. Computational theory proposes itself both as explanatory means and a direct method of creating consciousness. That makes it a more attractive theory to invest in than the others. First, because it hypothesis a means for generating consciousness and second, and quite crucially, those means are certainly within close reach. In fact, in the context of the discussion here (the Net’s possible state of consciousness), the Net itself could be a large and complex enough network to demonstrate the validity of the computational theory.
It May Already Be Conscious
One of the most notable features of the internet is not that any fool can use it, but that billions of fools can use it simultaneously. This is possible because there are billions of computers connected to each other simultaneously, a fact which is the internet.
However, without the fools the net would not exist. Nor (for the present at least) would it stand a chance of evolving further. In addition to developers and scientists, it is the human user that is at the moment driving the internet and its evolution.
But apparently that situation is not for long. The similarity between the complexity and connectivity found over the Internet and that of the human brain has drawn some predictable speculation. If you are a subscriber to the computational theory of mind, you might agree that it is only a matter of time before the internet becomes conscious, and will then one day be able to direct its own evolution.
As such it would be the brain and nervous system of a larger interconnected “social organism” consisting of both humans and technology. It is an idea envisaged by many researchers into the evolution of complexity, among them Dutch cognitive scientist Francis Heylighen. According to Heylighen,
Nobody knows what the future of humanity or society will be, and nobody can clearly imagine what an eventual global brain will look like… Yet, since it seems likely that something like a GB will evolve, whether we like it or not, it is worth trying to understand as much as possible about it, so that we may steer it in the best possible direction.
The computational theory attempts an account of mental processes using the principles of information processing. It’s central pillar is that mental processes and states (such as “pleasure”, “happiness”, “fear” etc.) are equivalent to, or at least arise from, neurons processing information in a manner that is similar to that of a computer. Clusters of jangling neurons (after having converted the incoming sensory data that provides the raw data for computation) are deemed to cause mental states that apparently account for sensations, awareness, meaning and ultimately beliefs and emotions. Fire a neuron or two; give birth to thought and feeling, is probably one way of describing the theory. As already mentioned, some see in it a means of explaining what consciousness, in the elaborate human sense, actually is.
The theory also crucially rests on the idea that any substance with a sufficient number of parts, and capable of performing vast computations, can produce a form of consciousness. Consciousness, the theory goes, will emerge from the parts and connections as they perform operations, almost as a by-product of function and almost, but not quite, by magic.
As mentioned earlier, some computationalists believe that some form of emergent consciousness is inevitable in any arrangement of functioning matter once a critical complexity is reached.
If one makes the observation that the Net is exceedingly large and complex; hosts perhaps trillions of dynamic web pages and software routines; is hosted and interacts on possibly a couple of billion computers; and is operated by an internet population of about 3 billion humans globally and rising, then, (Dr Frankenstein), we seem to have a potential case for a new form of consciousness. Provided you are a computationalist, the conditions for some kind of “global consciousness” seem to already exist in abundance.
Like many atheists, I’m open to the ideas of computational theory, as it offers the best alternative to traditional Cartesian dualism – the paradoxical idea that mind is a totally separate non-material entity that nevertheless somehow interacts with matter.
It is also quite plausible that if the theory is correct, the Internet is already conscious. Except we are unaware of it. This idea is in some way attractive, and the possibility does not unduly strike terror in one’s heart.
To arrive at the concept of a global brain Heylighen starts from the idea that society itself can already be viewed as a kind of super-organism. The functions of different organizations in society can be compared to the functions of organs, systems and processes in the body.
For example, industrial plants extract energy and building blocks from raw materials, just like the digestive system. Roads, railways and waterways transport these products from one part of the system to another one, just like the arteries and veins. Garbage dumps and sewage systems collect waste products, just like the colon and the bladder. The army and police protect the society against invaders and rogue elements, just like the immune system.
This analogy applies to society more or less as it has been for millennia, an idea that was not lost on the ancient Greeks as Heylighen points out. But though human brains direct this super-organism through governments, markets and in modern times media organisations and so on, never before have individual human brains been connected to each other so widely with such speed. According to Heylighen,
The main idea is that the different documents available on the web can be seen as nodes in a network, connected by a number of links, along which variable numbers of users travel in search for information.
In other words, web surfing. This is what constitutes the global brain claims Heylighen, a planetary ensemble of humans and hyperlinked documents. The documents are authored, searched and read by humans who are interlinked to each other through the documents themselves. And as the human authors change and add to the documents, both the documents and the humans change and evolve.
In addition to this, software algorithms designed to mine, collate and organise related and relevant data (under certain criteria) push the evolution along ever further. Often without the intervention of humans. The critical point is that such self-organisation is fundamentally recognised as the basis for intelligence as well as various forms of awareness and, ultimately, self-awareness:
[T]he technological revolution has produced a global communication network, which can be seen as a nervous system for this planetary being. As the computer network becomes more intelligent it starts to look more like a global brain or super-brain, with capabilities far surpassing those of individual people.
There are apparently already some pre-emergent properties on this human-electronic hybrid, consisting of programs that sift and transfer data, learn and re-learn how to self-organise and then deliver or store the data as necessary. The self-organising, self-learning is designed by people of course, but it is still self-organising.
Good, bad or just plain indifferent?
Even so, the intriguing aspect is not just whether the electronic part of the Internet-human hybrid, or “global brain” as Heylighen innocently calls it, will become aware in some way; but whether it will be “malevolent”, “benevolent” or perhaps indifferent to humans. Inevitably humans must always reduce the world to these attributes, since it is one of the ways in which we can extract an extra bit (or byte) of information from it, over and above that of raw manipulation of data. Conceptions of good and evil, apparently peculiar to humans alone, in particular provide an extra dimension for abstract processing of raw data, from which we then draw meaning (semantic content) that enables us to judge intentions. Meaning is our forte as conscious beings.
Teaming up against us
Although there is more reason to err on the side of benevolence and, much more likely, indifference, as always the more interesting possibility is the one that leads to mischief. What would a wicked, self-aware internet look like?
One possibility is that it would provide us with the means for a great deal of time wasting. Which is just the sort of thing you would expect from an entity that was preparing to overrun humanity. David Byrne of the Guardian speculates that the net may be conniving to keep us connected.
As the internet has become more integral to our lives, we’ve become more vulnerable to its seductions, and the web has started to act like a bully, a drug dealer. It knows we need it, love it and are addicted to it, so it can take advantage of that need.
The time-wasting charge is actually taken seriously by Susan Blackmore, a researcher into memes and consciousness and author of The Meme Machine.
More or less, this mischievous (as yet unproven) potential that the Net possesses for wasting our time allegedly arises from the fact that it is in the nature of self-organising systems to always organise themselves better for their own survival and growth. Which is selfishness by another name.
We know, for example, that our own original masters, the genes, evolved “survival machines” for their own use, machines that are intimately known to us as “our” bodies. Billions of years later the genes’ offspring, the information based memes, emerged, having evolved in the very brains that the genes had by then built for “their” bodies. (The question of ownership is still hotly debated). Memes brought with them the phenomenon that we call culture, and both they and genes have carried on evolving to greater heights of complexity ever since. We are at the point in their joint evolution in which we have modern human biological bodies and a stupendous informational and technological culture. Just who we are hasn’t been worked out yet, but let’s not be detained by that.
The two – genes and memes – simply get along just fine reproducing themselves selfishly in “our” bodies and brains. It’s something they do for their own benefit, completely unaware of either themselves or what has emerged from them, which is no less than little old us. Also, we know that the possibility of “our doing well out of the deal is but a mere accident of evolutionary design. So far we have existed purely at our masters’ unconscious pleasure, and it is only because our bodies and brains continue to provide genes and memes with an optimal place to reproduce that we have so far continued as a species.
So memes arose from genes (after which somehow out “we” popped from betwixt the twain). Blackmore uses an apt metaphor to describe this process of replicators begetting replicators, likening it to the myth of Pandora’s Box (actually a jar rather than box), in that each replicator holds a surprise (its offspring) that comes with significant consequences attached.
Cultural [memetic] evolution is a dangerous child for any species to let loose on its planet. By the time you realize what’s happening, the child is a toddler, up and causing havoc, and it’s too late to put it back. We humans are Earth’s Pandoran species. We’re the ones who let the second replicator [the meme] out of its box, and we can’t push it back in. We’re seeing the consequences all around us.[Ref]
But according to Blackmore there is now another little devil in the works, apart from us three (genes, memes and “ourselves”). She suggests that the second replicator, the meme (son of gene) has now given birth to a third, and claims this interloper is currently stalking the world with its own albeit unconscious agenda. This new replicator she calls a “teme”, with the t standing for technology. Just as the meme is characterised as the cultural child of biological evolution (that is, of genes), the teme she views as the technological child of culture (i.e., of memes). Whereas memes are defined as units of cultural information that breed in and jump from brain to brain, she defines temes as units of technological information that breed in and jump from machine to machine.
So the gene is now a grandfather if Blackmore is right. It evolved first. Then the meme. And finally we have the teme.
Booted Out of the Evolutionary Loop
The most important thing in what Blackmore is saying is not that there may be another replicator. It is her quite substantial claim that temes are binary data that is processed by machines only, rather than human brains. This third replicator has its own Darwinian path for survival and replication, and in following this path it entirely bypasses humans (the cheek of it!). This is a very significant step.
Replicators have stalked the Earth for eons in the form of the gene and meme, but until just a few decades ago, prior to digital computing, they were confined almost exclusively to our bodies and minds. It is true that memes have lurked for millennia in the artefacts we produce, such as art, music, tools, writing and so on. But they have no capacity to reproduce there, that is the crucial point. Memes till now have been entirely inert once they leave our minds, and so the rate of their evolution and diffusion among us was limited by the human population, which is itself limited by biology (the genes, again).
But now that memes can flourish in the new medium of software, and given that software can “beget” software, they can evolve all on their own. Countless programs now write other programs fully autonomously, and it is these programs that Blackmore calls temes, the third generation of replicator. And, it only lives and breeds in our computers and servers, as part of the programs and hypertext languages on our PCs and on the internet, and nowhere else. If that is really the case, then there is something else on earth that is evolving all by itself, largely independent of humans, but which intimately interacts with them and on which humans depend.
Blackmore cites apparently innocuous examples of temes and “teme complexes”, such as programs that write essays and poetry or ones that suggest books and clothes you might want to buy and places you might want to visit next. She also considers search engines as prime examples of the environment in which temes can evolve independently. Search engines copy, splice, select and then re-assemble information ready for presentation; frequently to us; but also, very frequently, to other programs and machines without ever meeting a human mind. This copying, selecting and re-assembling is none other than Darwinian evolution Blackmore claims, and her thesis seems to be this: binary information technology, though we have created it and let out of the box, is evolving by itself without the intervention of human brains. It may still be in its primitive stages, but so were genes and memes at one point, and just look at how ubiquitous and successful they have become. (A 2008 TED video talk Blackmore gave on memes and temes explains more on her idea, with interactive transcript allowing you to jump to keywords of interest.)
It is an interesting idea for a third replicator, but as with meme theory it lacks a fully empirical treatment that conclusively shows a true Darwinian evolution. Yet Blackmore may be right to caution us – “the Pandoran species” – with regard to the possible existence and uncontrolled release of potentially one more powerful and ubiquitous replicator, perhaps most recently coming at us, if we grant validity to the theory, in the form of Apps that integrate into every aspect of life via smart devices and the like. Some are very useful, but at what cost?
I have to admit to some initial confusion as to where precisely the dividing line is between a meme and a teme. Though, on reflection, it seems that Blackmore draws a distinction between the actual binary information that allegedly makes up units of culture and ideas – that is, memes; and the binary information that is used to manage those memes in electronic form. As I understand her, temes manipulated by machines control how memes are distributed and interact in our brains. Unlike memes, temes do not so much contain cultural information, but rather, simply manipulate it when it is in electronic form.
The teme theory resembles good food for thought at the very least, but it also raises again the question of the ultimate effect on humans of technology. If temes are real, how might we get the best benefit out of them without them benefiting more from us? Temes, if they are real, may not be any more good or noxious than memes or genes as a category or genus of replicator. But it is still necessary as well as interesting to ask what the effect of their existence might be on us.
Before considering that, there is one more observation to make about what researchers into the evolution of complexity say regarding the possible emergence (as opposed to the deliberate creation) of artificial intelligence over the Internet. People like Heylighen, who point to the possibility of some form of internet consciousness, do not necessarily claim that it may become conscious in the same way that we experience: that it will know that it exists and begin asking such questions as “why do I exist?”, “what is the purpose of compiling these search results?” or indeed such questions as “what is the purpose of humanity?” and so on. It could acquire human-like consciousness, who knows; especially if the computational theory is correct. (If consciousness is simply a question of ordered and sufficient complexity driven by computational processes, then as the internet-web ensemble enlarges, the possibility of “computational consciousness” increases).
What perhaps can be said about the claims people like Heylighen and Blackmore are making is that the Internet appears to have the potential to acquire characteristics and functions that to us may appear to be conscious. Or at least may appear as if they were brought about by a conscious agent (as suggested by Byrne). The Net would not have to actually be self-aware in the human sense for it to exhibit either intelligence or some level of awareness and control of its environment. After all, animals in general function on much the same terms, and some are pretty clever and devious at getting what they want, even though it is principally the higher mammals who appear to have a sense of self and purpose.
Perhaps the Net may actually become conscious in a human sense, and also know what it is like to be conscious and so on. But that does not matter with regard to the effect it may have. If any of its functions contain a degree of self-serving behaviour (basically, survival, growth and reproduction), then it does not matter whether it is conscious or not. All it has to do is behave as if it were conscious. And if Blackmore is right and Darwinian evolution is emerging over electronic communications and computing networks without our intervention, then there necessarily will be self-serving behaviour regardless.
It is this indifferent, morally innocuous consideration that is the most interesting. A tree that affects your house isn’t aware and cares even less that it is affecting your house. It simply grows according to its own (unintentional) evolved plan. It might undermine the foundations and cause you to die one night in your sleep, possibly the best way to die I would imagine. Or it might seemingly lovingly shade you from fierce sunlight and perhaps even provide you with fruit. But as a product of indifferent Darwinian evolution, whatever the effect it does produce, it will take what it requires to survive, grow and reproduce despite being void of consciousness.
Human and Internet: Co-evolution or Conflict?
The same might be true of the internet as for a living organism such as a tree or sheep. Whether conscious in the “I know I exist” way or not, if the Internet possesses the capacity to organise itself for greater efficiency (granted by us or just as an incidental by-product of what it is) it may likely use that organisation to at minimum ensure its survival and perhaps also grow and “reproduce”.
No malevolence is implied in this view. It would simply be the natural outcome of some kind of awareness or at least autonomous self-organisation directed by Darwinian evolution.
If that could be the case, could there ever be a conflict of interests between humanity and the Internet? Is there anything the Internet needs that humans also require, the pursuit of which could potentially cause trouble if ever the twain had to compete over it?
One obvious answer is energy. Apart from relatively little dedicated raw material for components and infrastructure, the Net mostly requires electrical energy to power it. If you account for the fact that the components themselves require energy for their manufacture, energy is really the only factor to consider.
That leads us to an interesting scenario. Given that both humans and the Net rely on energy for their survival, if the Net ever became “aware” of a future energy crisis, would we expect it to do the honourable thing and switch itself off? Or at least scale itself down in deference to its human masters?
That is probably a null question, because humans are now so reliant on the Net that downsizing itself to save energy would undermine the very economies that create the energy that nurtures it. An aware computer network would quite possibly foresee this dilemma before we did, and if it were to be sufficiently aware and “benevolent” or more likely indifferent, it might strive to find an optimal solution for our combined growth and consumption. Perhaps it might very well succeed where human economists have so far failed in terms of equity, but how would it do it, that is the question?
There is no good reason to think that it would ever become overtly malevolent over the matter. Despite our reliance on it, if it ever started picking us off by, say, fiddling with operating theatre equipment or traffic lights, we’d sooner pull the plug than risk bringing to reality a well-worn science fiction theme. Most likely it would not risk that, even if it in fact turns out to be “malevolent”.
It is more plausible that a conscious Internet would likely be quite rational and “amoral” over the issue of our co-existence. It would likely understand the intertwined nature of its existence and ours, and probably would strive to find an optimal co-evolution.
Yet though it might have this awareness, it may well fail to understand or appreciate the kinds of things that are important to humans. We can doubt that it would understand that quality of life as well as meaning and purpose are as important to us as straightforward existence and survival. It might therefore make the error of “thinking” that what is important to us is that which, one imagines, might be important to itself: which is perhaps simply to persist and perform its function.
It is difficult to think of any other conclusion an aware Net would draw when contemplating its own existence. Perhaps in its own grand scheme of life, to survive and perform one’s function is the supreme goal to pursue. If that were to be the case, then it might well be stupid enough to think that our function is to provide it with energy! This speculation remains plausible even if it paradoxically concluded that its own purpose in life is to function so as to serve humans! Indeed, it may well conclude that the supreme goal for both itself and us is to stay alive and working with the sole aim of – staying alive and working. Ad infinitum in a senseless and meaningless cycle. It might therefore reason that in order to perform its part in this circular and pointless modus operandi it needs to exist; and therefore we must provide it with energy, which is of course our part of the deal.
What is certain though, is that whatever the Net might conclude is the purpose of existence, energy will be paramount in its calculations. Perhaps then, its solution to any possible energy crisis would be to order the world such that we perform our function optimally in energy terms. So that we can provide it with energy… So that it can serve us…
It might do this, we can speculate, by causing us to refrain from using energy wastefully, in favour of expending it when performing functions that provide it with energy and growth. It may contrive ways of keeping us indoors and away from “frivolous” pursuits like motor racing, jet skiing and anything else that consumes energy without adding to what it might consider the principal function of its own and our existence (Jeremy Clarkson may like to ponder this).
And if it did allow us out every now and then, perhaps it would suggest destinations and activities that suited “the grand purpose” and not those which we desired.
Again, this does not mean that an internet aware in this way would be intentionally selfish or malevolent. It only suggests that on its view of existence it has concluded that survival and function are the reasons for existence. It may not know – how might it? (though indeed it might know) – about notions that require direct experience of millions of evolutionary years of adaptive trial and error to develop. We cannot tell it what meaning is, or what love or honour are, or at least it is difficult to imagine how it might be made to understand and appreciate such notions from a human perspective. Humans can themselves just about manage to learn them over a life time.
So more than likely its “ethical” values will be alien to us, if it were to have any, because whatever they may be will probably be related to and determined by what it is made of and what it does: wires, silicon transistors, nuts and bolts and heat sinks, plus vast amounts of energy and raw binary information processing. Notions of love or honour that a human would recognise may possibly evolve in all of that, eventually, who knows. Especially if it is true that all mental processes and psychological states that we would identify as emotional and value laden are at base “only” information processing, on the lines of the computational theory of mind. Feelings might well be “substrate neutral”, meaning that if a collection of silicon and wires were to become sufficiently aware it could possibly acquire a personality that humans would recognise and relate to. It is an idea that has been around for over half a century since Alan Turing proposed the famous “Turing test” used to test for human-like machine consciousness: if a machine can converse with a human without the human suspecting it to be a machine, who is to say that the machine isn’t conscious in the same way the human is?
But a more plausible speculation is that something quite strange would arise as equivalent qualities for perception and experience, as well as for emotions and beliefs. In fact, we might even doubt that its constituent matter would play a very large part in whatever “feelings” and “notions” it might evolve. It would be hard to envisage it experiencing pain when someone over tightens a bolt, or pleasure when someone wipes some dust off its heat sinks! Its “values”, “feelings” and “significances” are likely – initially at least – to be purely abstract, not involving sensory perceptions. They will probably be based only around naked mathematics, logic, statistics and so on, together with hard factual input. What a brute it might be, come to think of it.
But not so fast. Biological brains apparently use all of these processes, so we are calculating machines. Although as a matter of fact we have also consistently proved to be quite brutal calculating machines.
Yet, what appears to save us is the hitherto longevity of our species and our general evolutionary ancestry. Our brains have had millions of years to develop special and exquisite levels of abstraction, that do involve unique sensory inputs, and which lead to notions like love, humour, jealousy and everything else that makes life both a pleasure and a pain, and therefore ultimately meaningful and worthwhile. So in the end perhaps, though it is certainly not a forgone conclusion that a colossal non-biological machine could be brutal simply because it is perfectly rational, it is nevertheless also quite difficult to imagine how it might be humanly rational.
To illustrate this, consider how a cyber-dominated global brain might view humans born with disabilities and those who acquire them. Clearly, in productive terms it may view those of us less abled as a burden to economy. It might therefore, purely on logical, economic grounds, decide to “filter” out disabilities from the population when faced with a critical situation. That would be the rationalistic machine approach one would expect from pure cost-benefit calculations incapable of accounting for personhood and individualism.
That is in total contrast to the humanly rational approach. Most people recoil at any such consideration, and categorically reject any such calculation (though sadly not all). Generally, we are not interested in cost-benefit ratios, because the issue is not one of material but of moral value. Most will argue from the absolutist position of morality, that human life is sacred no matter its manifestation and capacities; others will argue from the deontological perspective. And a few, like a purely rational AI, may adopt the consequentialist. Generally though, most will uphold the notion that what we decide about others who are dependent on us does not simply and exclusively reflect what we think about them: it in fact also reflects what we think about ourselves as a conscious and ethical species. In other words, it reflects what the human race, able and less able, thinks about itself.
But perhaps more crucially than this, there is an absurdity to consider that would lead to our own self-destruction if humanity held to such rationalistic social engineering: those who turn on their fellows today, can expect to become victims tomorrow, should their fortunes change. And this does not even begin to touch on notions of love and delight that can exist among all humans whatever their condition.
We have arrived at the obvious really. How an aware cyber brain, whether a little android or the entire Internet, would relate to human value can be the only consideration worth worrying about.
Is the Internet Already Conscious?
Looking around the Internet, we might wonder as to whether Blackmore is right and the Net has already started to benevolently distract us.
How do we know that it isn’t already conscious in some meaningful way? When we perform a search, does it really dish up the most relevant content; or does it offer content that though it might be related to our search also holds a benefit for the Net itself? Might it offer up content that would cause us to remain in front of our monitors in preference to going out, when not at work, to save energy?
As Blackmore notes, software routines already suggest many things to the indolent surfer (that’s you and me), including where to shop or holiday and what films to watch. So who knows whether in amongst those suggestions are ones designed to serve the Net’s view of what is “right” and not only our view of what is right or what we want. It is rather funny to think that a search for information might perhaps return results that could cause you to change your belief on something, or to hold an opinion you never held before. This is in addition to the by now established “search bubble” within which search engines tend to place users. Search engines learn what and how a user searches for information, and then tailor search results to what “they think” the user is looking for. With the effect that users tend to have their beliefs confirmed, or at any rate only see content that they “agree” with and which is likely to be neither challenging nor novel. In effect, online, they exist in a “bubble”.
A case in point is the search engine Google. In July 2010 Google was taken to task (yet again) over how it conducts itself over the web. The organisation was accused of manipulating search results returned by its search engine such that services provided by Google are preferentially displayed at the top of the listings. [Arxive.org Ref.] Google denies any impropriety. However, given the possibility of temes and maybe even of internet consciousness, Google could take a stab at a novel defence: perhaps the humans at Google are not the culprits skewing the results and instead it is the Google bot, Google’s quasi-omniscient internet crawler, that may all of a sudden have discovered its purpose in life. Pardon us to be so mean, it was not us it were our temes – we might hear search engine providers cry one day.
But that is somewhat different to the Net as a whole contriving to direct humans towards behaviour ultimately designed for its own benefit. Or at least for our mutual benefit, by keeping itself evolving so that it can better serve us.
Finally, there is the malevolent path for the Net’s evolution. If it acquires consciousness sufficient to be aware of itself as an entity, it may discover that its continued evolution requires that humans are exploited, ruthlessly perhaps. The best way for it to achieve this would be to monitor our every move, much like the NSA and GCHQ have been doing for some years now. It may even be the case that the NSA and GCHQ have already been “infiltrated” by the Net for want of a better word, laying before them possibilities and tools for increasingly effective surveillance. Achieving social conformity by instilling an acceptance of a culture of surveillance in humans would provide an effective route through which the Net could obtain the economic security required to guarantee it unlimited energy. Thus, though the zealots at the NSA and GCHQ appear to love what they do, the Net may nonetheless also be nudging them along with suggestions on how to watch us more closely, in the same way as billions of Net users are nudged along with suggestions for films, shopping, recreation, recipes, politics and everything else that humans care about.
We have certainly made more than a few wild speculations here; though this might be forgiven as this is after all a book for dinner parties. What stands out as sound among the speculations is the energy consideration however. If one day there will be limitless energy, why would the Net bother itself with humanity? It may even deign to help us out a little. But if it comes down to an energy crunch time, and the Net knows it is energy crunch time, what would cause it be noble towards us, or at least not indifferent to our plight? It will be very interesting to see what the future holds for the co-evolution of humanity and the internet.
 Conscious self-awareness is likely to be an interplay between genes and memes. And if not these, then it will be some other information based real-world entity as yet undiscovered.
 Some research seems to indicate that even birds and spiders may have rudimentary selves. See Wild Justice: The Moral Lives of Animals, Marc Bekoff and Jessica Pierce, 2009.
 It is probably true that as the Net grows the amount of energy required will increase proportionately faster than the raw materials needed to build more infrastructure. One supposes this is true despite the fact that the components making up the infrastructure are becoming smaller and more energy efficient. The Net will probably still require increasing amounts of energy because it will perform more tasks, quicker.