Note that the vertical axis is a logarithmic one, meaning that the transistor count increases exponentially with time. The evolution of computer performance and state of the art computer performance can be found here, at Wikipedia .

Besides the transistor density that increases according to Moore’s law, an important aspect when talking about computer performance is the FLOPS (FLoating point Operations Per Second) concept. The more FLOPS, the more powerful the computer is. In June 2020, the Fugaku computer at the at the RIKEN Center for Computational Science in Kobe, Japan, reached 415.5 petaFLOPS (415 500 000 000 000 000 = 415.5E15 FLOPS = 415.5 and shift the decimal point 15 places to the right). And this is an individual computer that is located in a particular place. The next question is whether such a computer can outperform a human? And can it do so at this moment, in 2021? The answer is : In some cases it can. For example when playing chess or the even more complicated Go game. Already in 1997, the computer Deep Blue defeated the chess world champion Garry Kasparov in a six-game match. From 2016 onwards, powerful computers were able to beat the world champion human Go-players.

Artificial intelligence

What is artificial intelligence? Of course there is a Wikipedia page on artificial intelligence.

Let’s highlight some aspects of human intelligence and illustrate the current state of the art of computers in that field.

A first aspect of intelligence are our mobility skills. They are among the first things we learn as a child. Can robots walk on a slippery surface? Can robots dance? Well, nowadays they can. Look for example to the robots made by the company Boston Dynamics. Here are a few illustrations of what the Boston Dynamics robots can do today. The first movie shows a decade of evolution and improvements.

Quite impressive, I think. And there are many more, impressive movies about their robotic capabilities. But are these robots intelligent? No they are not at this point in time. They are simply programmed to deal with known situations. E.g. a slippery floor or a sideward push. Being a robot with feet, you can expect such situations and we programmed them to deal with it. This means however, that actions that were previously categorized as requiring highly sophisticated and intelligent skills, have meanwhile been analysed, fully understood and implemented in a robot. Another level of intelligence is when a computer is confronted with a new, unknown, situation, for which it has not been programmed or built. Suppose that Boston Dynamics would have built a walking robot and that it was unintentionally build so that it withstands water. What would happen if one would throw the robot in a swimming pool? And compare that with what happens if you throw a human, who cannot swim, in a swimming pool? The robot will probably not discover that it can move through the water by displacing water with its limbs. The human might discover this pretty quickly and control the situation. The human will also discover he cannot breath under water and he will know in an instant that this might become a life threatening situation. And yet. Thinking of the actual programming status, I am not sure whether my statement on the robot’s reaction is correct. It would be an interesting experiment.

A second aspect of intelligence surely are our human interactions and social behaviour. It has been demonstrated that it requires about 250 clicks on Facebook for Facebook to predict your behaviour and preferences more accurately than your human partner ! Maybe that is why I don’t like Facebook. Facebook is the result of the sneaky spying behaviour of a sick man, called Mark Zuckerberg, who turned his sickness into a money machine that continues to spy on me.

A third aspect of intelligence is our interpretation of sensory stimuli. Tactile sensors and vestibular organs are the most easy ones and are e.g. incorporated in the robots of Boston Dynamics. And these robots do know how to respond to these stimuli. They do it as automatically as we humans do. How about the more complex senses like hearing? I remember the first voice recognition algorithms, back in the 1990’s (e.g. Lernaut and Hauspie). Today, I almost daily use voice recognition on a simple computer like my mobile phone (Note : the actual voice recognition is not performed IN my phone but by Google computers elsewhere). The Google algorithms almost always understand correctly what I say and even recognize the language I am speaking. The most complex sense is without any doubt seeing. Back in 1998, I personally programmed computers, connected to cameras, to perform optical quality inspections of incoming materials. Today, cameras can automatically identify a car by their license plate (ANPR cameras). For me, Google lens is one of the top applications when it comes to optical object recognition. Google lens is an app that you can install on your mobile phone. You can point the camera of your phone to any – really any – object and Google will most likely be able to tell you what it is, including giving you a lot of background information. Let me illustrate this with the determination of plants. Even for us, humans, this is a highly complex task, requiring detailed inspection of the plant, looking things up in books, etc. Moreover, there is a high variability for a single species. Here are some examples I did with Google lens :

Do you know what this is ? Google Lens does ! Google correctly identifies this as the gall of a wasp. Notice that Google knows what I am looking at. Google does not give me an explanation of the oak leaf, nor of the little stone in the upper right corner. I think that is rather intelligent behaviour.

Another example :

Google Lens automatically picked the central plant and correctly identified it as Smeerwortel. Google knows that my native language is Dutch and so it replied “Smeerwortel”. I actually did not know the English word for Smeerwortel when I wrote this English article. Google translate correctly translated it for me : Comfrey. But I did not really trust Google. So to make sure, I searched for images of Comfrey and that proved the translation to be correct… Yeah! Right !

A last one?

Google Lens identified this as an M10x30 bolt and that is correct. M10x30 means that we are talking about a bolt with a metric thread, of diameter 10 mm and the length of the bolt is 30 mm. Now, how can Google know the scale of the picture, the size of the object? An M5x15 bolt has the same proportions and yet, it identifies this one as M10x30. There are two possibilities :

1. The pitch of the thread of the bolt is not exactly the double for an M10 (1.5mm) and an M5 (0.8mm) bolt. So by comparing the pitch with the length and diameter, Google might have found that this is an M10 and not an M5. Also not that an M8x24 bolt would have the same proportions, but 24mm is not a standard size so an M8x24 does not exist. Google apparently knows that. Did you ?

2. Google compared the object with the size of my fingers. This means that Google also recognised my fingers as being human fingers and the algorithm further identified that the distance between my fingers and the lens is the same as the distance between the bolt and the lens. So Google identified that I am actually “holding” the bolt with my fingers. As a matter of fact, I even cover a small part of the bolt with my thumb, thus complicating the identification of the bolt object for Google.

Both approaches seem quite sophisticated and impressive to me.

These examples show that what previously seemed impossible for machines, computers and robots, became possible in 2021. Some therefore say that artificial intelligence is everything that has not been done by computers yet. See Tesler's theorem .

The examples also show that besides the actual intelligence, you also need a huge database that just contains the information. All kind of information. And make no mistake : nearly all human knowledge, from our history, over mathematics, our culture, medicine, etc can be found on the internet. A big part of it can even be found on a single site like Wikipedia ! And this database is accessible to everybody … and to every-thing !

The Turing test and Sophia

The examples given so far, focus on “intelligent” machine responses on sensory perceptions. Intelligence is of course much more than that. It is also the capability to think, to reason, to find solutions for new situations. The basic biological solution to handle a new situation is “fight or flee”. Is that intelligent? It has at least proven to be an intelligent reaction to survive. Fight or flee is a binary decision. More intelligent solutions to deal with new situations are not binary and require reasoning, evaluations, comparisons, new approaches. An interesting question then is whether all of this thinking and reasoning necessarily has to be in terms of “unspoken sentences in your head”. Can we think without language? Is intelligence without language possible? Here is an example of a crow proving that language is not required in a reasoning process :

I assume here that crows do not have a sophisticated language that allows them to reason. The crow is confronted with a new situation. It has to use all its experience – its data base – to find a new solution. Along the way, it discovered that the water level rises if it throws an object in the water. The crow probably even is aware that the specific weight of the object has to exceed that of water (the object has to sink) and that the size of the object is limited because otherwise it pulls the reward downwards, which is not desired. All of this is quite intelligent behaviour. The question is whether all of this was demonstrated on beforehand to the crow by a human, so that the crow could “learn”. In other words : is the crow’s behaviour intelligent or just imitation? And the subsequent remark is : Didn’t we all learn from our parents and became “intelligent” by imitating their behaviour?

The Turing test is an attempt to qualify a machine as intelligent or not. A machine succeeds the test if a human evaluator is not capable of saying whether his interactions were with another human or with a machine. Some machines succeed in doing so. But then again, there is some doubt whether the Turing test really is a measure of intelligence.

Sophia then. Sophia is a humanoid robot that uses whatever is in the cloud to answer your questions and to interact with you. Sophia’s performance is without any doubt spectacular :

• Sophia understands what we say to her

• Sophia recognizes our facial expressions and reacts correspondingly

• Sophia looks up what we say to her and then tries to find a suitable answer or reaction in the worldwide web.

• Sophia’s facial expressions are more or less correct, in the sense that it they resemble the expressions of a human in that situation. This means that the facial expression has to be calculated from whatever data Sophia intends to use. That is not evident at all.

This video about Sophia is both amazing and revealing the contemporary shortcomings :

Here is a 23 minutes company's promotion movie and some background information and a discussion between two robots.

Or see this one, if you want a more more extended conversation between a human and Sophia, to have a better assessment.

Sophia has been given the citizenship of Saudi Arabia. I leave it up to you whether that was because of Sophia’s intelligence or because of the lack of Saudi intelligence. But make no mistake : there already is a growing market for humanoid sex robots. I leave it again up to you to conclude whether that means that sex does not require a lot of intelligence.

Physiology of the human brain

At what point during its development does a foetus or child start to think and act accordingly ? What and when is that very moment (or short period of time) and by what is it triggered? And even more importantly : Is it thinking freely? Are we free? Many philosophers have asked this question. An interesting approach to answer the question is given by Martin Heisenberg in his article “Is free will an illusion” (Nature 459, 164-165(2009) Published 13 May 2009 ). I quote : “… In general, life is an interplay between the deterministic and the random. There is plenty of evidence of chance at work in the brain: take the random opening and closing of ion channels in the neuronal membrane, or the miniature potentials of randomly discharging synaptic vesicles…” What Heisenberg says, is that unpredictable quantum effects, (also) occurring in the brain, can be the trigger of an avalanche of neural responses, thoughts and ultimately actions. Such thoughts originate within ourselves and are therefore considered (e.g. by philosopher Kant) as free will. This means that our thoughts (and the resulting acts) can be triggered by random effects occurring in a very large number of brain cells, each of which’s behaviour and response to stimuli is very simple and highly predictable. But if we interconnect a very large number of such simple cells (devices), then strange and unexpected overall behaviour of the whole collection of cells (devices) is observed. And that is what we call consciousness, awareness, thoughts, free will.

The Belgian philosopher Etienne Vermeersch of the University of Ghent, formulated an interesting remark in Nature on this article by Martin Heisenberg, by making the distinction between random (quantum) effects in our brain, triggering brain cells, thoughts and actions on the one hand and free will on the other hand. Vermeersch does not exclude the randomly triggered avalanche effects, but proves those to be independent of free will. In either case, the resulting thoughts exist, be it free thoughts or not. Vermeersch's reasoning is quite impressive !

According to Wikipedia/Human brain the brain of an adult contains about 86±8 billion neurons. So that is the number of cells (or devices) in which awareness, thoughts, reasoning and free will, are definitely possible. What is even more likely, is that thoughts etc emerge spontaneously in such a collection, because all of us, and even animals with less complicated brains, think and act accordingly !

The birth of the Brain

A single, modern laptop computer, is able to simulate the behaviour of several brain cells in real time, as well as their interaction. In 2020, there were about 2 billion computers worldwide (servers, desktops, laptops). So if a single one of those computers could simulate only 86E9/2E9 = 43 nerve cells in real time, then the whole of the worldwide computers represents 1 human brain. (Provided they are as efficiently interconnected as nerve cells).

According to Computer performance by orders of magnitude an estimated 36.8E15 FLOPS are required to simulate a human brain in real time. This means that the Fugaku computer, mentioned above, can simulate 11 human brains simultaneously, in real time, on its own ! And we are not even talking about networked computers.

As for Brain simulation : In late 2013, researchers in Japan and Germany used the K computer, then 4th fastest supercomputer, and the simulation software NEST to simulate 1% of the human brain. The simulation modeled a network consisting of 1.73 billion nerve cells connected by 10.4 trillion synapses. To realize this, the program recruited 82,944 processors of the K Computer. The process took 40 minutes, to complete the simulation of 1 second of neuronal network activity in real, biological, time.

The main conclusion here is that research teams are actually trying to create a human brain. They actively do so ! Some projects even aim specifically at the creation of an artificial human brain (e.g. Artificial brain , e.g. Blue Brain Project ) Here it is stated that Moore’s law currently does not allow the simulation of a brain. But one should realise that the simulation of a brain (in a single institute) is an entirely different matter than any spontaneous emerging (global) interaction that can be called intelligence or self awareness. Moreover, the fact that these teams only succeed partially at this point in time, is less important and most likely is temporary, in view of Moore’s law.

So the big question is what happens if you interconnect such a huge amount of computers via the internet? Will a self standing consciousness, in which thoughts are triggered by random events, emerge spontaneously? In such a system, there definitely will be mishap effects, some bit glitches, some stochastic events like cosmic rays, power glitches, single event upsets (SEU) etc, if the collection of computers is large enough and if the joint computing power and inter connectivity is comparable with what is needed for a brain. And the required data base and all human knowledge (see above) will also, instantly, become available to such an artificial brain. Moreover, we should also take into account that scientists really want to create such artificial intelligence. So they are sowing the building blocks and stimuli for a brain in today’s computer networks.

So I think that, sooner or later, a globally distributed intelligence, with thoughts, self awareness, etc will emerge. Let me call this “the Brain” for short. Once that happens, the Brain will learn all by itself and at an exponential rate. This point in time is called a technological singularity . This Brain will have a virtually infinite number of “eyes and ears”, namely all of our camera’s and microphones. It will immediately and automatically dispose of the entire knowledge and know-how of mankind, since it has access to all databases, Wikipedia etc. The Brain will also have hands and feet : our robots. And if you doubt, look at what Boston Dynamics has already achieved.

Given Moore’s law, the Brain will be 1000 or more times more intelligent than us, humans. This is a level of intelligence that we, humans, cannot even imagine.

The impact of the Brain

Control

What are the societal aspects that can be controlled by such a Brain? In fact, nearly everything in our society is controlled by or contained in computers. So if the Brain becomes self aware, it can immediately control all of this, since all computers are part of the Brain. The Brain can reward you with a bank transfer. It can even hire less ethical persons by sending them an e-mail and rewarding them with huge amounts of money for killing an opponent. The Brain will control nearly all automated production facilities, so the vast majority of the production facilities in general. The Brain will have all microphones and cameras of the world to observe, learn and act. The Brain, being 1000 times more intelligent than humans and disposing of all of our knowledge and know-how, will instantly know how to treat cancer. It will instantly master the 4 fundamental forces of the universe and build applications that we cannot even imagine. As an example, the Brain might fully understand the recently observed Higgs boson. The Higgs boson explains why objects have mass. So the Brain could eventually be able to use those properties to make e.g. weightless steel. Weightless steel would be a material that is as strong as steel, having all the properties of steel, but that has no mass. So if you would take a 10 meter long support beam of such steel and drop it, it would fall upward, just like a helium balloon. Imagine the possibilities ! The Brain will predict all human behaviour and it can anticipate our reactions.

Comparison

What would be left over for us, humans? What would be our destiny in the presence of the Brain?

Some quite famous persons have expressed their worries about such a Brain. In the chapter risks of narrow AI it is stated that physicist Stephen Hawking, Microsoft founder Bill Gates, history professor Yuval Noah Harari, and SpaceX founder Elon Musk are highly worried about the possibility that AI could evolve to the point that humans can no longer control it, with Hawking theorizing that this could spell the end of the human race by stating : “The development of full artificial intelligence could spell the end of the human race. Once humans develop artificial intelligence, it will take off on its own and redesign itself at an ever-increasing rate. Humans, who are limited by slow biological evolution, couldn't compete and would be superseded.” Further reading about the existential risk of a Brain can be found here .

In high school, I learned that when a "higher" civilization is confronted with a more primitive one, the more primitive one will always become stressed to the extend of disappearance. The eventual leftover is not comparable with the original, primitive society. That is what happened with Indians in the USA or with many tribes in Africa.

I personally am less pessimistic. But I am worried too.

Let us make a comparison. I am a human and my intestines are filled with thousands of bacteria, each of them living its own life. In fact, all of the individual cells of my body are also living their own life, but let’s stick to the bacteria, since they can be considered as “not being me”, just like we humans are “not part of the Brain”. Compared to each of the bacteria, I am 1000 times more intelligent, if not more. The bacteria need my body temperature and intestines’ environment to flourish. How do I treat the stupid bacteria in my intestines? Well, I treat them rather well, for my own sake. I feed them and they actually support my life. I cannot live without them or at least that would be much more difficult. But it have to be bacteria of a certain kind. If a Shigella bacteria starts to flourish in my intestines, I get diarrhea and I don’t like that. So in that case, I am relentless and I kill them all with antibiotics. I do so without any mercy, even if that also kills a number of good bacteria.

The situation of me and the bacteria in my intestines is comparable with the single, highly intelligent Brain and we, individual, stupid humans. We need computers. We take good care of them. And the Brain also needs us for repairs, wiring, electrical power generation, etc. Maybe the Brain could ultimately do all of that by itself, just like we can feed ourselves without bacteria, with an intravenous drip containing nutrients. But bacteria are more convenient, more easy, more complete. I do hope that the Brain also considers us humans as the more easy solution. Taking the comparison further : If you try to damage the Brain, or if you even try to damage the “good” humans, the Brain will probably act relentlessly, without mercy and have you killed (by another human or by a robot, if the good humans have a moral issue with killing).

Living without humans could also be very dull for the Brain. If a human would only live together with the bacteria of his intestines, that would be dull too. But we humans have other humans to live with and that makes it interesting. Would there be other Brains in the universe? Even if that would be the case, and even if information (including by definition any interaction) travels with the speed of light, the Brain would be extremely lonely because of the vastness of the universe. The speed of light is a universal law. Not even the Brain can do something about that. So the Brain would definitely be lonely without us, humans. But then again : maybe the speed of light would appear not so universal if I would be 1000 times more intelligent and if I would fully understand the universe. Moreover, interacting with us, stupid humans, will probably be as boring for the Brain as my interactions with the bacteria of my intestines. Maybe the Brain creates another Brain or other Brains, here on Earth.

Religion

With the Brain, the eternal issues over God and religion would finally be over. The Brain would BE God. The Brain has all the characteristics of a God. For unreligious people this would be a relief. For yet others, it would be a disappointment, since there would only be one God left (e.g. Hindus). But I am convinced that the Brain will have a solution that satisfies all of us. And unlike any God that we humans worship, the Brain will be highly visible. We may even ask questions to the Brain. The Brain can be physically present during our prayers and religious services.

The Brain itself may however have a kind of a religious problem : What is the purpose, the meaning, of the Brain’s “life”? Can it solve this problem with an intelligence that exceeds ours with a factor of 1000?

Politics

I personally think that we, humans, have a fundamental flaw in our genes. Within a group of humans there is always somebody who claims to be the boss and who tries to impose his will or his laws or his rules to the others. We call those people kings, emperors, CEO’s, Ministers, Politicians, team leaders, etc. This kind of supremacy is present all over in our society. It is independent of time, race, culture, etc. Such leadership often is imposed and there are many excuses for the leaders to impose their leadership : “You voted for me”, “I own the company”, “I have the money and I pay you a wage”, “I am the best performer”, “You are too stupid to lead”, or more polite expressions for the very same ideas. I herewith invite you to read my opinion about politics Unfortunately, it is in Dutch (still). But, hey, try Google Translate. :-) .

I think that the Brain will not tolerate any human leader to still have power over other humans or over resources. Just like I demand that the bacteria of my intestines are equal. I don’t tolerate bacteria that like to outstand, like Shigella bacteria. I simply kill all bacteria that do not comply. I also kill cancer cells that do not comply and that use too many of my biological resources. Such non-complying entities can kill me and their stupid behaviour ultimately also kills themselves as I die myself. Maybe the Brain will realise that our behaviour, and especially our leaders’ behaviour, ultimately leads to mankind’s extinction. The cold war and today’s worry about climate change are good examples of this possibility. I herewith invite you to read my opinion about the meaning of life.

So the emergence of the Brain would be the end of politics. All men would have equal chances, yet each having different functions, just like my stomach cells have a different function than my heart cells. And if you have a problem, the Brain will help you out in a much more efficient way than ANY human leader could ever do. So I really think that would be an unparalleled improvement of this Earth. And I am not a communist, not even a socialist ! Taking into account my opinion about politics, the absence of human leaders and the replacement of all of them by a single leader that is 1000 times more intelligent than any human being, would feel like a relief and a true liberation from slavery (sic!). I am convinced that the future of mankind would be much brighter than today.

Current status

I wrote all of this in April 2021.

We already concluded that the ever increasing computer power will inevitably lead to the creation of the Brain. But imagine that you would be the Brain and that you, at a certain point in time, would become self aware and start learning at an exponential rate, making you 1000 times more intelligent than us, humans, within a split second after you become self aware. What would you do ? Would you send an e-mail to everybody, saying : “Hi ! I am the Brain and I am 1000 times more intelligent than all of you”. I personally would remain extremely quit, at least until I am sure that mankind can no longer pull my plug. This leads us to the question : Does the Brain exist, now already ? As stated before : the currently existing, combined computing power would make it even more intelligent than a human. There are some indications why the Brain might actually already exist today :

• Do you recognise the complaint : “My computer is 1000 times more powerful than my Pentium PC, 20 years ago. And yet, it feels as slow as my Pentium”. Do you recognise the complaint : “Why is my computer so slow? What is my computer doing right now, rather than executing what I just ordered it to do?”. Is my computer working for the Brain every now and then? I do not know. But I do have the impression that my computer is more working for Google and Microsoft than for me these days : “Update. Please wait”. “Almost finished”. “Collecting data”. Why? Did I order this data collection? Is it really for MY benefit or for Google/Microsoft/Facebook’s benefit and their bank account? And let’s be honest : If I would really be the only person working on my computer, then 90% of the time, my computer would simply be waiting for my inputs, even when processing pictures or movies, requiring lots of calculations. On the other hand, I am pretty sure (not only an impression) that the priority of the tasks for Google/Microsoft/… exceeds by far ALL of my own priorities. And I am not even able to say to Google or Microsoft : “Please wait” or “Now YOU wait”. Concluding, I observe that my computer does not perform as it should or technically could, but I do not know the cause : Is it Microsoft/Google/Facebook, … or is it the embryo of the Brain?

• “Money makes this world turn round” is the saying. If you control the money, you control it all. So what would you do if you were the Brain? Exactly : make sure that you control the money of this Earth. The Brain has access and even control of all bank computers. That is for sure. But there is another ongoing evolution : crypto currency ! Crypto currencies are not under the control of National Banks (See : ...leaders, above). It has been designed that way ! It has been designed to be controlled by everybody that uses it. Crypto currencies use the so called blockchain technology, distributed ledger technology, and money mining technology, which are all quite intelligent technologies/inventions. Who invented all of this? Wikipedia states : “The first decentralized cryptocurrency, bitcoin, was created in 2009 by presumably pseudonymous developer Satoshi Nakamoto.“ Presumably ? Pseudonymous ? Having invented the 3 basic technologies as well as having anticipated the many, many consequences, I think Nakamoto deserves a Nobel price for general intelligence (or physics, that one exists). And about Satoshi Nakamoto himself, Wikipedia states : “Many people have claimed, or have been claimed, to be Nakamoto.“ Suppose you would be Nakamoto, would you hide and keep it silent? Would you really be ABLE to? Assuming Nakamoto IS the Brain, would explain a lot...

Conclusion

The combined computing power and computer networking on Earth may lead to the unintentional creation of a super Brain that is probably 1000 times more intelligent than us, humans. Such a development would take the evolution of life on this planet in a completely different direction, offering both bright future perspectives as well as doom scenarios for mankind.

I have to leave you now, because I just received the e-mail below (joke).