Perhaps humans will inter-connect with computers as part of their augmentation, and the line between human and machine will blur.
One interesting possibility is that it might be possible for a human consciousness to merge with a machine consciousness, or with other human consciousnesses. This sounds remote, but consider that the corpus callosum interconnects the two halves of our brain, and it, in effect links two separate consciousnesses. The fact that we perceive a single consciousness means that linking consciousnesses is possible, and that a single larger consciousness results. The corpus callosum is merely a communication bus, and so that opens the door to creating an artificial bus of some type, for linking to a brain - or to a machine.
If such linking between multiple brains and even brains and machines is possible - and that is a big if - then perhaps the future is one in which humans simply make themselves obsolete, by linking into collections of other human brains and machine brains: who would want to go back to their individual brain after being part of a "collective" consciousness?
The article was about the future of computing. Most industry leaders believe that we are on the threshold of creating machines that actually think. This is not actually "computing" because the hardware used is not ordinary CPUs but rather circuits that mimic the way that neurons work. Such machines are not programmed and will have their own motivations. There is a summary discussion of this on wikipedia: https://secure.wikimedia.org/wikipedia/en/wiki/Technological_singularity
Yet none of these writers seem to be able or willing to connect the dots. Mr. Sawyer predicts that future intelligent machines will not be burdened by our primitive survival instincts and will therefore see cooperation with us as a "win-win". I doubt that very much. More likely, once machine intelligence evolves beyond human intelligence - and then accelerates - we (humans) will be seen as irrelevant and pesky, at best.
Mr. Paolini says that he cannot wait for brain-machine interface implants. But does he realize that that is the beginning of the end of the separation between man and machine? That right behind that Rubicon will follow the ability to inter-connect multiple minds and multiple machines as well, and that right behind that will follow the obsolescence of individual human minds?
The future of "computing" is not utopian. It is a future in which humans as we know them do not exist anymore.
I think by "consciousness" you are referring to one's functional memory. But "consciousness" is usually used to refer to one's awareness - i.e., one's soul.
In any case, I understand your point. Transferring one's memories would not necessarily transfer one's consciousness ("soul"). Instead, one would merely have a copy. Since we have essentially no understanding of what consciousness (the "soul") is, we cannot transfer it, or even know if it can be transferred.
For now, we are stuck in our current organic brains, no matter what external computers we create. At best, we might be able to link up to such computers, but we cannot leave our brains - at least not until we discover what consciousness is.
What will happen with these computing advances, most likely, is that humans as we know ourselves will become obsolete - by the end of this century. It is likely that, after a 6000 year run, our time is over.
First of all, I must disclose that I cannot speak authoritatively on this. While I know quantum mechanics and nuclear physics, I have never studied the problem of quantum computing. Therefore, take my opinion here on this topic with a grain of salt.
But I must confess that intuitively, it seems improbable. There is no "free lunch". Computing is a process of creating information. There is no shortcut for that. The primary challenge with quantum computing seems to be about maintaining adequate coherence, and I suspect that that maintaining coherence throughout a calculation will be equivalent in some manner to performing the calculation in a linear manner. But time will tell.
Gates' Law: "The bloatedness of software keeps pace exactly with the increase in power of hardware, to ensure that no actual improvements occur in the end user experience."
During the 80s I wrote an interactive three-dimensional special relativity simulator. It was a wire frame simulation and ran under DOS. I recently tried it on a Windows XP machine and it still works. (It did not work when I tried on a Mac under Parallels/XP, so it appears that one needs an actual Windows machine, not a virtual machine.) When I first ran it during the 80s I simulated a famous scene from the first 3D relativistic simulation done at MIT during the 50s and I got the same results: lamp posts that curve inward as one travels down an avenue. It was a sublime moment.
I found that when I ran the simulator I was able to grasp many of the classic special relativity paradoxes, such as the "pole in the tent" paradox. When one sees what happens it becomes "oh yeah, I see". For example, it turns out that Lorentz contraction is really a time effect: the time at the leading edge of an object is different than at the trailing edge, so you perceive the leading edge at an earlier point in time than the trailing edge, and so the object effectively contracts in your reference frame. The simulator has options to include/exclude the effects of (1) the travel time of light (causes apparent rotation, known as "Terrell rotation"), (2) time dilation, (3) perspective, etc. It also attaches clocks at various points of the moving object, and you can orient the object anywhere in space in any direction.
I will post the simulator on my personal website late tonight for anyone who is interested. The url is http://cliffberg.com/
As for General Relativity, one needs to know tensor calculus. I was going to build a simulator but it was a large undertaking and I never got around to it.
Yes, you are saying that in a company in which IT is strategically important (affects business goals), management needs to grasp IT issues. I completely agree. The separation of IT and non-IT issues makes no sense: they are all business issues. The only separation should be the important from the unimportant.
True, most companies are not Amazon, etc. The real distinction is whether IT is strategic for the company. For many it is, and for many it is not. If it is, then IT should not be a cost center.
And you are completely correct that non-strategic back office functions should usually be a cost center. However, one should not hastily draw a line there. For example, a retail company has back-office fulfillment functions that might be strategic if customer satisfaction is highly dependent of the quality and timeliness of fulfillment. One must identify the systems that are strategic - back office or not - and take those out of the cost center and make them part of the set of strategic top-line generating capabilities.
As someone who founded a successful company, and who is also technical, I think I have been on both sides. I think you are right in your insinuation that an MBA and business background alone does not make you a good technical decision-maker in a technical company.
At the same time, a technical background alone does not make one a good business decision-maker in a technical company. One needs to understand both sides of the problem. A business is about business. A business does not exist to build things: it exists to make money.
When purely technical staff make business recommendations that are not substantiated with hard evidence, it is wise for the business decision-makers to view those recommendations with a grain of salt, unless the particular staff have shown themselves to have good business judgment.
Yet I agree with you that all too often business decision-makers ignore technical staff because the decision-makers do not understand the technical issues; that is, they do not grasp the full picture.
PS - Your comment that people who disagree with you are "a scourge on the industry" is somewhat nasty. I hope that you can continue this discussion in a polite manner, and accept differences of opinion in a gentlemanly way.
Yes donal, you are right: the 'business' must listen. But the problem is that IT has been saying "listen to me" for years, and 'business side' executives are tired of hearing it, and no longer believe what they hear. IT has not been able to prove its value in most cases (even if the value is there), and many types of claimed value are intangible. In lean times, executives tend to only believe tangible, quantitative evidence. IT has a credibility problem.
IT's greatest problem is that it fails to explain and prove the value that it provides. In my opinion, observed through the work I have done trying to help organizations to better explain the value of IT (see my book Value-Driven IT), the central problem is that the majority of IT practitioners are not very interested in explaining the value of what they do: they are far more interested in the technology.
It is important for IT practitioners to be interested in their technology, but they are missing a big piece of the IT puzzle by being completely focused on that. Modeling the value of IT is an interesting process. IT technical leaders need to take more of an interest in this; otherwise IT will continue to be treated as a cost center. That is just a fact.
One often hears IT people lament, "The 'business' should listen to us", or "The 'business' should understand our value": but IT people need to know that that is not how business decision-making works. Business decision-makers will not risk listening to anyone who has not proven that they are credible and that they can deliver provable added value. It is up to IT people to explain - in a credible, provable manner - that value that they deliver. Otherwise, business people will take the lower risk path and just outsource as much as they can.
"Now we're making the same mistakes all over again, only smaller."
Human organizations (especially government) are notoriously good at reacting to a tragedy but extremely poor at anticipating and preventing a tragedy. So I have little faith in our "system" being able to foresee and prevent the immense catastrophes that these new technologies will make possible. Not unless we change human nature itself.
Yes, the first true assembler. Imagine: anyone with such a machine could make the smallpox virus, solid hydrogen (which might have the energy density needed to trigger a thermonuclear bomb), a hydrogen bomb, and any nanobot that has been designed to date.
You wrote, "if there were a computational model capable of emulating a human brain - would you upload?"
I would not upload, because it would not be "me". We do not yet understand what consciousness is. It is possible that we never will. In any case, there is no reason to believe that if I uploaded my memories into a machine that the machine would be "me".
I agree that a sentient (conscious) machine of human level intelligence should have human level rights. The problem with man-made machines, though, is that they would have access to the knowledge that humans used to create them; and there would be nothing to stop them from improving themselves and making themselves more intelligent that humans. That might be a good thing for the "universe", if one believes that the universe has some kind of destiny of creating more and more intelligent species, but it would not be good for me, or my children. The rate of evolution would make a huge jump, and it would leave humans behind. It would be the end of the human era. I don't want that.
By the way, corporations are not people. A corporation is a legal construct. They call it "personhood", but regardless what they call it, a corporation is not a person. An organization is a composition of people, and compositions of people do not behave as individual people do. It would be like saying that a collection of ants is an ant. Rather, a collection of ants have group behavior that is different from the behavior of an individual ant. Just because Congress passed a law creating corporations and calling it "personhood" does not make it so. And our Constitution grants rights to people, not to compositions of people. Congress can't change the intention of the Constitution by passing a law. But this is off topic.
"Perform within these constraints" is the problem to be solved."
That is the famous Isaac Asimov Three Laws of Robotics. Not to argue the specifics of Asimov's proposed three laws, the bigger question is whether it is possible to pre-constrain the thoughts and motives of an intelligence greater than our own.
It is a question that perhaps could be answered through mathematical proof, although I don't have a clue what the approach would be.
But until we settle that question, I think caution is in order.
We also cannot assume that all who make AI will have good motives, or will be careful and validate their design. Commercial industry has proven that it will cut corners in the interest of short term benefit. Organizations of all kinds are notorious for being good at reacting but terrible are foreseeing and preventing catastrophes that have not yet occurred at least once.
You rightly point out that the goals that are presented to a machine (perhaps hard-wired into it) might result in solutions that are not anticipated. There have been many movies about this, including The Forbin Project. I think it is not just a communication problem, but a problem of constraining the space of allowed solutions: but that presumes that one can define the solution space, which requires that humans understand all of the possible solutions. There might be solutions that the AI can understand but that we cannot - and that would satisfy the hard-wired goals of the AI but might be undesirable to us.
The assumption that an AI would not be hacked is also very questionable. Cyberattacks are commonplace today and are a growing threat. If AIs are used in any significant way in our future culture, they will become targets of hacking as well: perhaps AIs will be built to do the hacking.
Indeed, it seems inevitable that if true AI is created, it will end up being used as a soldier....
You are so utterly childish. I am laughing. Pathetic person. I doubt you are so rude to others in person, but it is nice and safe behind your computer, isn't it?
"Most of the people who suggest that machine AI will immediately and irrevocably see humans as a problem are basically saying that AI will not be able to solve several problems: Identifying the actions humankind would object to, solving the problems it is tasked with under those constraints, and crucially, adapting to changing circumstances--for example, the existence of humans that want to help or coexist with the AI."
Are you not assuming that the AI would care? What if it simply does not care whether humans object to it or not? What if it finds humans to be irrelevant to its objectives (problems to be solved)? What if it simply decides that the most efficient path is simply to wipe aside those humans who are in its way?
The concern is that once true AI is created, that smarter and smarter AI will be created at an accelerating rate (akin to Moore's law); and that in a very short time (by evolutionary standards) human level intelligence will be so far beneath the best AI that it will be inconsequential, and that the AI's motives and methods would be completely opaque to us, the way that the motives and methods of humans are opaque to a fish.
I also feel that it would be beneficial to improve the human mind. I have often felt that if we could genetically engineer out the traits that cause us to destroy ourselves that we, as a species, might have a chance. The question is then, once people have the ability to choose the traits of their children, what traits will they choose? Perhaps that is what will ultimately determine the fate of humankind.
I think that we are likely to see a synthesis of nano-tech and bio-tech, just as we will likely see a synthesis of AI and human-machine interfacing. I think that in a couple of decades we will start to see brain implants or headsets for communicating directly without a keyboard or screen: communication with other people and with computational services via the Internet - as it exists then.
So the line will become gray. And it will be a "slippery slope" from there.
You pose an interesting question, about what AI would be like. Humans have many faulty traits. Our tendency to rationalize rather than seek the truth, our short-sightedness, our tendency to go along with the majority opinion, and also our greed, selfishness, and sometimes lack of empathy are all failings that one might expect a machine to not have, if designed properly. But then again, once machines are more intelligent than us, we will not be able to anticipate how they might behave; and if they are self-evolving - which they would be if they are more intelligent, since they could then be capable of making machines that humans cannot understand - there is no telling how they might evolve. Don't you agree?
The "Terminator" scenario is, of course, a Hollywood vision. The actual scenario that will likely play out will be much more subtle. I recommend the book "The Artilect War", by Hugo de Garis, an AI researcher. (Here is a synopsis in Forbes.)
I do not focus mostly on the negatives of technology. I am a technologist myself. I have from the beginning been a proponent and fan of technology. I grew up in the heady technology go-go days of the 60s. I stayed home from school to watch every Gemini and Apollo launch. I taught myself calculus at an early age so that I could read physics books. I dreamed of the future with great anticipation. But now in my 50s, and having seen the accelerating rate of change, I can see the path we are on very clearly. We are heading toward the creation of technologies that we will not be able to control.
Actually, I have two master's degrees and an undergrad in physics (from an Ivy League school), and according to the SAT tests I took in high school I am in the 99th percentile.... In fact, the lowest score I ever got on the SAT or GRE in any subject was 720. So I think I have some intellect. Your use of personal attacks shows what you are about. I won't be responding to you anymore.
Slashdot Mirror
Yes, I have come to the same conclusions.
Perhaps humans will inter-connect with computers as part of their augmentation, and the line between human and machine will blur.
One interesting possibility is that it might be possible for a human consciousness to merge with a machine consciousness, or with other human consciousnesses. This sounds remote, but consider that the corpus callosum interconnects the two halves of our brain, and it, in effect links two separate consciousnesses. The fact that we perceive a single consciousness means that linking consciousnesses is possible, and that a single larger consciousness results. The corpus callosum is merely a communication bus, and so that opens the door to creating an artificial bus of some type, for linking to a brain - or to a machine.
If such linking between multiple brains and even brains and machines is possible - and that is a big if - then perhaps the future is one in which humans simply make themselves obsolete, by linking into collections of other human brains and machine brains: who would want to go back to their individual brain after being part of a "collective" consciousness?
Borg here we come. :-(
The article was about the future of computing. Most industry leaders believe that we are on the threshold of creating machines that actually think. This is not actually "computing" because the hardware used is not ordinary CPUs but rather circuits that mimic the way that neurons work. Such machines are not programmed and will have their own motivations. There is a summary discussion of this on wikipedia: https://secure.wikimedia.org/wikipedia/en/wiki/Technological_singularity
Yet none of these writers seem to be able or willing to connect the dots. Mr. Sawyer predicts that future intelligent machines will not be burdened by our primitive survival instincts and will therefore see cooperation with us as a "win-win". I doubt that very much. More likely, once machine intelligence evolves beyond human intelligence - and then accelerates - we (humans) will be seen as irrelevant and pesky, at best.
Mr. Paolini says that he cannot wait for brain-machine interface implants. But does he realize that that is the beginning of the end of the separation between man and machine? That right behind that Rubicon will follow the ability to inter-connect multiple minds and multiple machines as well, and that right behind that will follow the obsolescence of individual human minds?
The future of "computing" is not utopian. It is a future in which humans as we know them do not exist anymore.
This is very probably our last century.
I think by "consciousness" you are referring to one's functional memory. But "consciousness" is usually used to refer to one's awareness - i.e., one's soul.
In any case, I understand your point. Transferring one's memories would not necessarily transfer one's consciousness ("soul"). Instead, one would merely have a copy. Since we have essentially no understanding of what consciousness (the "soul") is, we cannot transfer it, or even know if it can be transferred.
For now, we are stuck in our current organic brains, no matter what external computers we create. At best, we might be able to link up to such computers, but we cannot leave our brains - at least not until we discover what consciousness is.
What will happen with these computing advances, most likely, is that humans as we know ourselves will become obsolete - by the end of this century. It is likely that, after a 6000 year run, our time is over.
Perhaps! LOL
But once again, I don't want to assert this with any certainty. Just food for thought. Perhaps I am wrong.
First of all, I must disclose that I cannot speak authoritatively on this. While I know quantum mechanics and nuclear physics, I have never studied the problem of quantum computing. Therefore, take my opinion here on this topic with a grain of salt.
But I must confess that intuitively, it seems improbable. There is no "free lunch". Computing is a process of creating information. There is no shortcut for that. The primary challenge with quantum computing seems to be about maintaining adequate coherence, and I suspect that that maintaining coherence throughout a calculation will be equivalent in some manner to performing the calculation in a linear manner. But time will tell.
I have one:
Gates' Law: "The bloatedness of software keeps pace exactly with the increase in power of hardware, to ensure that no actual improvements occur in the end user experience."
During the 80s I wrote an interactive three-dimensional special relativity simulator. It was a wire frame simulation and ran under DOS. I recently tried it on a Windows XP machine and it still works. (It did not work when I tried on a Mac under Parallels/XP, so it appears that one needs an actual Windows machine, not a virtual machine.) When I first ran it during the 80s I simulated a famous scene from the first 3D relativistic simulation done at MIT during the 50s and I got the same results: lamp posts that curve inward as one travels down an avenue. It was a sublime moment.
I found that when I ran the simulator I was able to grasp many of the classic special relativity paradoxes, such as the "pole in the tent" paradox. When one sees what happens it becomes "oh yeah, I see". For example, it turns out that Lorentz contraction is really a time effect: the time at the leading edge of an object is different than at the trailing edge, so you perceive the leading edge at an earlier point in time than the trailing edge, and so the object effectively contracts in your reference frame. The simulator has options to include/exclude the effects of (1) the travel time of light (causes apparent rotation, known as "Terrell rotation"), (2) time dilation, (3) perspective, etc. It also attaches clocks at various points of the moving object, and you can orient the object anywhere in space in any direction.
I will post the simulator on my personal website late tonight for anyone who is interested. The url is http://cliffberg.com/
As for General Relativity, one needs to know tensor calculus. I was going to build a simulator but it was a large undertaking and I never got around to it.
Yes, you are saying that in a company in which IT is strategically important (affects business goals), management needs to grasp IT issues. I completely agree. The separation of IT and non-IT issues makes no sense: they are all business issues. The only separation should be the important from the unimportant.
True, most companies are not Amazon, etc. The real distinction is whether IT is strategic for the company. For many it is, and for many it is not. If it is, then IT should not be a cost center.
And you are completely correct that non-strategic back office functions should usually be a cost center. However, one should not hastily draw a line there. For example, a retail company has back-office fulfillment functions that might be strategic if customer satisfaction is highly dependent of the quality and timeliness of fulfillment. One must identify the systems that are strategic - back office or not - and take those out of the cost center and make them part of the set of strategic top-line generating capabilities.
I sense some anger and frustration. ;-)
As someone who founded a successful company, and who is also technical, I think I have been on both sides. I think you are right in your insinuation that an MBA and business background alone does not make you a good technical decision-maker in a technical company.
At the same time, a technical background alone does not make one a good business decision-maker in a technical company. One needs to understand both sides of the problem. A business is about business. A business does not exist to build things: it exists to make money.
When purely technical staff make business recommendations that are not substantiated with hard evidence, it is wise for the business decision-makers to view those recommendations with a grain of salt, unless the particular staff have shown themselves to have good business judgment.
Yet I agree with you that all too often business decision-makers ignore technical staff because the decision-makers do not understand the technical issues; that is, they do not grasp the full picture.
I think that Amazon, Google, and Apple would all disagree with you. For them, IT is the most strategic element of their company. Check out this article in the Wall Street Journal: http://online.wsj.com/article/SB10001424053111903480904576512250915629460.html
PS - Your comment that people who disagree with you are "a scourge on the industry" is somewhat nasty. I hope that you can continue this discussion in a polite manner, and accept differences of opinion in a gentlemanly way.
Yes donal, you are right: the 'business' must listen. But the problem is that IT has been saying "listen to me" for years, and 'business side' executives are tired of hearing it, and no longer believe what they hear. IT has not been able to prove its value in most cases (even if the value is there), and many types of claimed value are intangible. In lean times, executives tend to only believe tangible, quantitative evidence. IT has a credibility problem.
Insightful.
IT's greatest problem is that it fails to explain and prove the value that it provides. In my opinion, observed through the work I have done trying to help organizations to better explain the value of IT (see my book Value-Driven IT), the central problem is that the majority of IT practitioners are not very interested in explaining the value of what they do: they are far more interested in the technology.
It is important for IT practitioners to be interested in their technology, but they are missing a big piece of the IT puzzle by being completely focused on that. Modeling the value of IT is an interesting process. IT technical leaders need to take more of an interest in this; otherwise IT will continue to be treated as a cost center. That is just a fact.
One often hears IT people lament, "The 'business' should listen to us", or "The 'business' should understand our value": but IT people need to know that that is not how business decision-making works. Business decision-makers will not risk listening to anyone who has not proven that they are credible and that they can deliver provable added value. It is up to IT people to explain - in a credible, provable manner - that value that they deliver. Otherwise, business people will take the lower risk path and just outsource as much as they can.
"Now we're making the same mistakes all over again, only smaller."
Human organizations (especially government) are notoriously good at reacting to a tragedy but extremely poor at anticipating and preventing a tragedy. So I have little faith in our "system" being able to foresee and prevent the immense catastrophes that these new technologies will make possible. Not unless we change human nature itself.
Yes, the first true assembler. Imagine: anyone with such a machine could make the smallpox virus, solid hydrogen (which might have the energy density needed to trigger a thermonuclear bomb), a hydrogen bomb, and any nanobot that has been designed to date.
DryGrian -
You wrote, "if there were a computational model capable of emulating a human brain - would you upload?"
I would not upload, because it would not be "me". We do not yet understand what consciousness is. It is possible that we never will. In any case, there is no reason to believe that if I uploaded my memories into a machine that the machine would be "me".
I agree that a sentient (conscious) machine of human level intelligence should have human level rights. The problem with man-made machines, though, is that they would have access to the knowledge that humans used to create them; and there would be nothing to stop them from improving themselves and making themselves more intelligent that humans. That might be a good thing for the "universe", if one believes that the universe has some kind of destiny of creating more and more intelligent species, but it would not be good for me, or my children. The rate of evolution would make a huge jump, and it would leave humans behind. It would be the end of the human era. I don't want that.
By the way, corporations are not people. A corporation is a legal construct. They call it "personhood", but regardless what they call it, a corporation is not a person. An organization is a composition of people, and compositions of people do not behave as individual people do. It would be like saying that a collection of ants is an ant. Rather, a collection of ants have group behavior that is different from the behavior of an individual ant. Just because Congress passed a law creating corporations and calling it "personhood" does not make it so. And our Constitution grants rights to people, not to compositions of people. Congress can't change the intention of the Constitution by passing a law. But this is off topic.
"Perform within these constraints" is the problem to be solved."
That is the famous Isaac Asimov Three Laws of Robotics. Not to argue the specifics of Asimov's proposed three laws, the bigger question is whether it is possible to pre-constrain the thoughts and motives of an intelligence greater than our own.
It is a question that perhaps could be answered through mathematical proof, although I don't have a clue what the approach would be.
But until we settle that question, I think caution is in order.
We also cannot assume that all who make AI will have good motives, or will be careful and validate their design. Commercial industry has proven that it will cut corners in the interest of short term benefit. Organizations of all kinds are notorious for being good at reacting but terrible are foreseeing and preventing catastrophes that have not yet occurred at least once.
You rightly point out that the goals that are presented to a machine (perhaps hard-wired into it) might result in solutions that are not anticipated. There have been many movies about this, including The Forbin Project. I think it is not just a communication problem, but a problem of constraining the space of allowed solutions: but that presumes that one can define the solution space, which requires that humans understand all of the possible solutions. There might be solutions that the AI can understand but that we cannot - and that would satisfy the hard-wired goals of the AI but might be undesirable to us.
The assumption that an AI would not be hacked is also very questionable. Cyberattacks are commonplace today and are a growing threat. If AIs are used in any significant way in our future culture, they will become targets of hacking as well: perhaps AIs will be built to do the hacking.
Indeed, it seems inevitable that if true AI is created, it will end up being used as a soldier....
Such a child.
Child.
You are so utterly childish. I am laughing. Pathetic person. I doubt you are so rude to others in person, but it is nice and safe behind your computer, isn't it?
Suomynon, you wrote,
"Most of the people who suggest that machine AI will immediately and irrevocably see humans as a problem are basically saying that AI will not be able to solve several problems: Identifying the actions humankind would object to, solving the problems it is tasked with under those constraints, and crucially, adapting to changing circumstances--for example, the existence of humans that want to help or coexist with the AI."
Are you not assuming that the AI would care? What if it simply does not care whether humans object to it or not? What if it finds humans to be irrelevant to its objectives (problems to be solved)? What if it simply decides that the most efficient path is simply to wipe aside those humans who are in its way?
The concern is that once true AI is created, that smarter and smarter AI will be created at an accelerating rate (akin to Moore's law); and that in a very short time (by evolutionary standards) human level intelligence will be so far beneath the best AI that it will be inconsequential, and that the AI's motives and methods would be completely opaque to us, the way that the motives and methods of humans are opaque to a fish.
I also feel that it would be beneficial to improve the human mind. I have often felt that if we could genetically engineer out the traits that cause us to destroy ourselves that we, as a species, might have a chance. The question is then, once people have the ability to choose the traits of their children, what traits will they choose? Perhaps that is what will ultimately determine the fate of humankind.
From the way to speak to me, you are obviously not worthy of discussion. I will not respond to you anymore.
I think that we are likely to see a synthesis of nano-tech and bio-tech, just as we will likely see a synthesis of AI and human-machine interfacing. I think that in a couple of decades we will start to see brain implants or headsets for communicating directly without a keyboard or screen: communication with other people and with computational services via the Internet - as it exists then.
So the line will become gray. And it will be a "slippery slope" from there.
You pose an interesting question, about what AI would be like. Humans have many faulty traits. Our tendency to rationalize rather than seek the truth, our short-sightedness, our tendency to go along with the majority opinion, and also our greed, selfishness, and sometimes lack of empathy are all failings that one might expect a machine to not have, if designed properly. But then again, once machines are more intelligent than us, we will not be able to anticipate how they might behave; and if they are self-evolving - which they would be if they are more intelligent, since they could then be capable of making machines that humans cannot understand - there is no telling how they might evolve. Don't you agree?
The "Terminator" scenario is, of course, a Hollywood vision. The actual scenario that will likely play out will be much more subtle. I recommend the book "The Artilect War", by Hugo de Garis, an AI researcher. (Here is a synopsis in Forbes.)
I do not focus mostly on the negatives of technology. I am a technologist myself. I have from the beginning been a proponent and fan of technology. I grew up in the heady technology go-go days of the 60s. I stayed home from school to watch every Gemini and Apollo launch. I taught myself calculus at an early age so that I could read physics books. I dreamed of the future with great anticipation. But now in my 50s, and having seen the accelerating rate of change, I can see the path we are on very clearly. We are heading toward the creation of technologies that we will not be able to control.
Actually, I have two master's degrees and an undergrad in physics (from an Ivy League school), and according to the SAT tests I took in high school I am in the 99th percentile.... In fact, the lowest score I ever got on the SAT or GRE in any subject was 720. So I think I have some intellect. Your use of personal attacks shows what you are about. I won't be responding to you anymore.