When some of the first computers were designed, many were shocked by their processing speed and mathematical capabilities. Researchers in the field of computer science were later fascinated by another conception, Artificial Intelligence (AI). They figured if a computer has the power to calculate numbers and retrieve various forms of information with such speed; why can't it think and operate like a human?
In the 1950s, the first AI conference was held at Dartmouth University. The conference allowed different computer scientists to share views on AI and its implications for the future. They believed that AI would be the next step in computer science. Many computer scientists used laws such as Moore's Law, to conclude that AI would drastically improve over a span of time. Moore's Law, explains how technological progression in exponential, contrary to being linear. The law also explains how microchips become smaller and faster, as time passes.
As more time and effort was invested into AI, many researchers were finding that the science of AI was becoming vastly complex. The notion of an AI acting, thinking and functioning like a human, was becoming unfeasible. Researchers were beginning to think that maybe a conventional computer cannot operate with substantial AI, equivalent to that of the human brain. Researchers then moved AI into a more biological field; they called upon The Defense Advanced Research Projects Agency (DARPA) to help solve the predicaments that AI faced. DARPA faced an unprecedented challenge, to reverse-engineer the human brain. In order to reverse-engineer the human brain, one would have to take apart the entire human brain. Mind you, the human brain is made up of trillions of cells. Furthermore, the human brain is the most sophisticated entity in the known universe. Was this task really possible?
The future of AI success then became entirely in the hands of government funded research. The Department of Defense was the primary source of funding for AI by the 1960s. By the 1970s, DARPA had announced that it wanted to keep attempting to reverse-engineer the human brain and critics of AI research wanted government cuts for AI. Once this funding was cut, the "AI winter" had officially begun.
By the 1980s, commercial computers were advancing considerably and home-based video games consoles were becoming popular, which sparked a renewed interest in AI once again. Towards the end of the 1980s, AI research once again became lethargic and a second "AI winter" occurred.
The 1990s, opened up new ideas for AI and allowed it to grow drastically. AI was being used for oil and gas drilling, logistics, medicine and mining. In 1997, the first computer defeated a human in a game of chess. Furthermore, in 2011 a computer defeated some of the top contestants of Jeopardy!
While all of these accomplishments are extraordinary, AI still did not resemble a human's intelligence and/or its full capabilities. AI than branched out into other areas, such as pattern recognition, representation, common sense knowledge, learning from experience, planning, ontology, heuristics, semantics and epistemology. Once again, DARPA had attempted to reverse-engineer the human brain, but failed.
In order for scientists to create an AI with human-like abilities, reverse-engineering of the human brain must be accomplished. According to inventor and futurist Ray Kurzweil, he believes this will be completed by 2029. Once this goal is achieved, scientists and researchers will fully understand how the brain entirely functions. Furthermore, this will permit AI to progress to a new level. In the past, Kurzweil had predicted many technological events exceptionally accurately. Kurzweil predicted the time frame that a computer will defeat a human in a game of chess and the processing speeds of supercomputers reaching unprecedented levels. Kurzweil also invented the scanner, omni-font optical character recognition, print-to-speech reading for the blind, music synthesizers capable of recreating a piano, large-vocabulary speech recognition and the text-to-speech synthesizer.
Once scientists reverse-engineer the human brain, a major paradigm shift will occur. We will then approach the technological singularity. The technological singularity is when a human being merges with a computer. In the end, this would create a Cyborg (Cybernetic Organism). A cyborg will be part-human and part-machine. It would incorporate technologies such as Mind Uploading, the Internet and Neuroprosthetics. There is a good chance that some people will be in favor of cyborgs and some may be opposed to them. People already use technology such as bluetooths, cell-phones and laptops. Some would argue; are we already becoming cyborgs? According to Kurzweil, all humans will become cyborgs and our lives will become easier and more efficient. We will have the capacity to merge technologies such as Nanotechnology into our computer and biological components. In addition, Kurzweil believes that the use of AI, will aid humans in everything such as entertainment, medicine, construction, manufacturing, mining and space exploration.
Kurzweil's views, created some controversy in the field of AI and computer science. Some experts in AI, openly challenge his viewpoints. They believe there is no such thing as "Friendly AI". Hugo de Garis, who has a PhD in AI and other degrees in computer science and theoretical physics, does not believe Kurzweils's perception of AI. De Garis, a builder of artificial brains, strongly believes that AI will eventually have computing power trillions of times more powerful than the human brain. Yes, even if the principles of Moore's Law start to slow down in the future. At these levels, he believes that AI will not be friendly to humans.
In his book, "The Artilect War: Cosmists vs. Terrans: A Bitter Controversy Concerning Whether Humanity Should Build Godlike Massively Intelligent Machines", de Garis explains how a potential conflict will occur between two human groups. These groups are the Terrans (humans that will fight the AI-builders and their supporters) and the Cosmists (the builders and supporters of AI-machines). De Garis does a superb job in explaining why and how humanity will eventually face this dilemma.
Although the concept of AI becoming more intelligent than humans may seem hard to grasp, theoretically speaking, it is entirely possible. In his book, de Garis uses the term "Artilect" representing an Artificial Intellect. He claims that an Artilect can and will be created with capabilities, which are trillions of times more intelligent than a human. However, this could take decades or even a century or two to accomplish.
The question many ask is; how can an Artilect become so intelligent? The answer is based on the amount of knowledge that computers will retain in the future. Think about it like this, when a person uses the internet today, is it superior to a human? For instance, Google uses a syntax computer program. When a person types in one letter, the program can find billions of potential matches. As you type in more letters the matches start to decrease, until the word you want pops ups, this is a form of AI. When you want to find information quickly; do you run to your computer?
In de Garis's books and lectures, he mentions many emerging technologies that will aid in building an Artilect. He focuses on three different technologies. One technology is reversible computing. Currently, we have irreversible-computing, which means a laptop, TV and other electronics give off heat. Reversible computing will be a colossal technological breakthrough, because it will allow three dimensional circuitry. Imagine a laptop that does not need a fan, it could run for decades or centuries, without ever overheating. Reversible-computing will be the backbone of an Artilect.
Another awe-inspiring technology will be topological quantum computing. This differs from a traditional computer because it can only process one program at once. Conversely, a topological quantum computer can process gargantuan amounts of programs at once. This would increase the overall yield of a computer extraordinarily.
Lastly, de Garis goes into detail about nanotechnology. It is a technology which can manipulate matter on a molecular and atomic scale. Nanotechnology can be applied to molecular biology, micro-fabrication, organic chemistry and other areas. This technology will allow super-small computers to be created. These computers can then reproduce themselves rapidly. Furthermore, de Garis mentions more advanced levels of nanotechnology which include picotechnology and femtotechnology. As these technologies advance, they become smaller and smaller.
There is no doubt that all of these technologies are achievable and probably will be created in the future. There is still one major problem with de Garis's concept. Can and Artilect really become self-aware? The science of self-awareness is where de Garis's opponents challenge him again. They believe that self-awareness is an impossible task to accomplish. How can an Artilect become its own conscience being? This is an area of science that has not been touched yet; is it really possible? According to de Garis, it is possible. This may be entirely possible, although chances of an Artilect becoming self-aware, may ultimately be achieved in an unorthodox manner.
For example, if a nation's military were to merge with AI, theoretically it would become self-aware. This AI/Artilect would do anything in its power to defend its nation. A major question still arises though; does it have ethics? What if it entirely disagrees with its generals and/or commanders and attacks the humans it works for instead? Once self aware, it may be superior to humans and decide to have its own military objectives. An AI/Artilect would have immense knowledge about military history and tactics compared to any human, even if they obtained a PhD in warfare and military tactics. Even though this is a potential scenario, de Garis still firmly believes that an AI/Artilect will be a self conscious being. His chief theory is that an Artilect will eventually wipe out humanity. He goes into great detail about species domination and evolutionary engineering.
Species domination and evolutionary engineering are the basic foundations of Darwinism and evolution. The idea is based on how creatures survive based on intelligence and skills. Around 150,000 years ago, another species of humans exited. These creatures, fought with our species of humans. Can you guess which species of humans survived? Yes, we did. How did we do it? We simply annihilated all of them. Our intelligence and skills were superior to these creatures, therefore we were the victors.
De Garis's concept of an Artilect wiping out humanity, clearly follows the principles of evolutionary engineering and Darwinism. If an Artilect had intelligence that was trillions of times more advanced than the capabilities of a human; how would we survive? Remember, humans have to sleep, eat, drink, use the bathroom, mate, and unfortunately get sick. A human must reproduce offspring by implementing a tedious process, which takes 9 months. What does and Artilect do all day? Two things, think and reproduce, in just a fraction of a second. Could the human race really keep up with them??? Furthermore, we have to look at all of the knowledge that is accessible today. We can go to a search engine and find anything, computers are already totally superior to humans. What will the internet and/or and encyclopedia be capable of in a few decades? How much more powerful will a simple computer be compared to the human brain?
The idea of an Artilect turning on humans, may sound like something out of a movie. In theory, it is entirely possible. De garis is a firm believer of this event occurring, he may be right. The biggest obstacle that humans face is the access of information that an Artilect will have. An Artilect can look back at basic history and view all of the wars that humans fought. They will wonder; why were humans so barbaric? Furthermore, they will look at the vast extrapolation of human deaths as wars progressed. Artilects could then clash with humans, simply because we disagree with them or just simply get in their way. Humans could just become a major nuisance to these superior beings. Artilects can use many weapons against humans, such as depriving humans of food and water or creating a deadly disease that spreads easily.
On the other hand, we are currently superior to many things around us as well, things such as bugs and bacteria. We kill them without even thinking about it. How can we be so cruel? Easily, it is because they are inferior to us. If you walk down the street, do you ever realize that you killed bacteria and bugs? The average person does not, but you sure do. What if an Artilect treated us humans like a bug or bacteria? One could argue, that humans would be in deep trouble.
Another area of interest concerning Artilects is the future of a Cosmist and space exploration. A Terran would be against the building of Artilects. Remember, what does Terran really mean? It deals directly with earth. So would a Terran really be concerned about space exploration? Probably not. On the other hand, a Cosmist would. So, what if a Cosmist focused so heavily on space exploration that they aided the Artilects in colonizing space? Let's move into the realm of astronomy for a minute. Our solar system is in the Milky Way. It takes billions of light years to travel through our Milky Way galaxy. How long is one light year? Give up? Try about 6 trillion miles. So, how many galaxies exist out there in space? At least 150 billion different ones. In theory, a Cosmist can build an Artilect that can colonize other celestial objects and just simply, leave earth. They can then keep expanding to other celestial objects. Humans would then be isolated on earth, another terrible predicament for us. An Artilect then would have the option to simply decimate earth as we know it, remember they can travel to any other planet and/or celestial object in the known universe. Humans have attempted to stay in space several times throughout history. Furthermore, humans have tried many times colonize other celestial objects and have failed in colonizing anything beyond the moon.
The concept of an Artilect thinking about astronomy and space exploration is captivating. It is clearly a hypothetical scenario, but how could humans really survive in space for long durations of time? An Artilect would be able to stay in space for months or even years on end. They eventually would be smart enough to modify propulsion technology, transportation routes, building/maintaining facilities and implementing In-situ resource utilization (ISRU). All of these technologies would allow an Artilect to make gargantuan advancements in the field of space exploration. Furthermore, Artilects could colonize not just distant moons or planets, but eventually other solar systems or even galaxies.
If humans fail to keep up with the Artilects; could humans just become a small footnote in history?
People such as de Garis are very concerned scientists. We all should recognize and thank him for his tedious research. De Garis mentioned that his hero was Leo Szilard. Do you know who he is? He helped develop the atomic bomb. Szilard was the man who warned the world about such a destructive weapon in 1933, but at the same time he was afraid that the Nazis could use this weapon against the allies in World War II. Szilard was one of the architects behind the Manhattan Project. Ultimately, Szilard wrote a letter to President Roosevelt warning him about a potential Nazi atomic bomb. Szilard later told President Roosevelt to show the Japanese the yield of the bomb, but not to actually drop the weapon on a civilian population. Szilard eventually wrote the "Szilard Petition", a document signed by 155 scientists warning the President not to use the atomic bomb on any civilian population. President Roosevelt then died and the Szilard Petition never reached President Harry S. Truman.
So, de Garis can be considered a modern-day Szilard. He is building artificial brains, although they can potentially wipe us out. Many would say; why would one build such a potentially dangerous machine? Well, why did Szilard build the atomic bomb? It is part of human nature. We are destined to strive and move forward in science. If we look further into weapons created, later on the hydrogen bomb was built. This bomb is roughly one thousand times more powerful than the atomic bomb. Who would build such a destructive device? Different governments decided to build them for military purposes. Some believe that governments may be the primary source of funding for AI in the future. Wouldn't a government want to have their military backed by AI? Moreover, not just their soldiers and weapons, but their overall command structure. Many suggest that advanced AI may be created by governments, just like nuclear weapons were.
So, were there any good outcomes of Szilard's ideas for the Manhattan Project? Of course, nuclear power! Could there be positive spin-offs from AI, of course. Science is simply just a double-edged sword. Technology has always been both good and bad. Take fire for example, it could be used to make a fire and cook. You can also stay warm near a fire. On the other hand, if you walk into fire, you will be incinerated and die a painful death. New technologies all work the same way.
So, can there really be an exceedingly intelligent, merciless, femtoteched machine that is fully self-aware? If you follow Moore's Law, yes it is totally possible. Computers get faster ever day, the internet doubles in speed every year. The term is called Bit-Rate Annual Doubling (BRAD). So yes, the internet will be trillions of times faster than it is now in just another decade!
After digesting all of this information; you should take the time to decide if are you a Terran or a Cosmist? If you are a Terran you and your family members might be fighting the Cosmists in the near future. The Cosmists will see you and your family as a bunch of old-fashioned earth loving humans. The Cosmists will not understand why you do not want to embrace Moore's Law and create a new species that will dominate the future. However, you do have one more option, become a Cyborg. You will then be forced to submit to the Cosmists, Cyborgs will still follow the principles of Moore's Law.
We should all view both intellects de Garis and Kurzweil as true heroes. They have dedicated so much of their time to predict what the future will bring. The debate between both of them will become more and more relevant as we move further into the future. This topic may seem bizarre and even silly to some, but it is extraordinarily crucial. We should all take de Garis's warning very seriously and hope that Kurzweil is correct in the end.
Selena Gomez ariel winter Paige Butcher David Petraeus Petraeus Mia Love wall street journal
No comments:
Post a Comment
Note: Only a member of this blog may post a comment.