The basic idea behind AGI research, is that biological brains are already an empirical example of general intelligence [See Wikipedia/general intelligence], that nature has produced, and ai researchers seek to replicate this already existent type of intelligence, in the form of human level artificial intelligence, also called artificial general intelligence.
Typically, we find that religious people fear artificial intelligence, labeling it as the "mark of the beast".
See: [Religious fear of human level artificial intelligence, on youtube, "Mark of the Beast"]
Subscription Note:
Choosing to subscribe to this topic will automatically register you for email notifications for comments and updates on this thread.
Email notifications will be sent out daily by default unless specified otherwise on your account which you can edit by going to your userpage here and clicking on the subscriptions tab.
I don't fear it because it shall not occur. Remember, all the bullshit you have been shoveling is to point to computers that only do what they are programmed to do. Nothing more.
rmfr
Citation needed.
No citation needed. If you need to have it explained to you that computers only do what they are programmed to do, then you are only continually proving you are not a STEM PhD candidate working on computer simulated intelligence.
rmfr
The ai algorithms start out stupid, then they learn how to do the task, without being programmed.
There are even some algorithms, like AlphaGo zero that don't require human data to do well.
~
Example: AlphaGo Zero Learns To Play Go From Scratch With No Human Data
It's quite clear by your words, that you likely don't do artificial intelligence research.
~
The reason why I asked you for citations, is to encourage you to minimize your ignorance by doing research. However, it seems you fancy pussyfooting around, either as some measure of trolling or some similar endeavour.
Alpha go zero was built from alpha go. Still required human programming.
What University are you studying at?
Avant Brown: "The reason why I asked you for citations, is to encourage you to minimize your ignorance by doing research. However, it seems you fancy pussyfooting around, either as some measure of trolling or some similar endeavour."
I do not need citations. I was studying into computer scienc and the possibiliy of AI before you ever itched your daddy's loins. As I have said, I am still pulling knowledge I have already learned in that field without citations. You cannot even formulate a simple sentence without having to resort to Wikipedia for 90% of everything you say. Guess that shows how much a volcanologist can alread know compared to a modern day STEM PhD candidate. Hmm...
I do not need to do anymore research other than just keeping up with the, what is it by now, the ExaFLOPs capabilities at LANL. I have a friend who works there. And LANL has one of the greatest computational facilities on the world. Yet, with that much computational power, even their "SI" software engineers cannot do any better than basically a machine that runs programs written by humans. NOTHING MORE. Your Holy and Pefect God is still nowhere near being created by humans. In fact, I'll have been dead of old age long before we even get close.
This bullshit you keep proselytizing as a New Age Religious Absolutist Apologist is just that. BULLSHIT! Machines still do not learn. Nor do they think. The epitome of intelligence. Yes, they can store data and peruse it to adjust with the task they are to complete. In a sense, this is learning. However, they are still machines only running programs written by humans to operate within a certain constraint as applied by humans. Those machines still do not truly learn. They still do not think.
Using your Go analogy, yes, the computer was still doing nothing more than "brute forcing" a pattern of moves in order to achieve the task of winning. You say the total number of possible moves in Go is greater than the number of atoms in the universe. So what? The computer farms at LANL can still brute force that number within what, a few seconds. Even at 1 ExaFLOP power, how long would it take to brute force through 1 × 10^600 possible moves? Exa = 1 × 10^18. Only about ½minute. And if that Go computer was using PetaFLOPs (as I have heard), then it could still brute force that many in about 40 seconds.
However, continue discussing your asinine proselytizing.
rmfr
You really seem to be quite ignorant thereafter.
As I pointed out in another response, AlphaGo is not "brute forcing" the search space.
That's because the search space exceeds the number of atoms in the known universe, and so one would need a computer the size of the universe to iterate through these possibilities.
I dont fear AI but I resent the planned widespread deployment for it by minority group of vested interests in areas affecting employment. I am an atheist and probably a trogladyte as well.
Do we really need cars to drive themselves?
I anticipate this motor vehicle fatality rate in the US for example, will somewhat answer your question.
Fatality and injury rates from standard cars aside, the US way over burdened highway system desperately needs all cars on the road to be self driven and communicating with each other.
If every car on the road was self driving, highway capacity would go up 3x 4x. And we could turn all our parking lots into parks or more homes for the cities desperately short in housing/space. Plus all of those millions upon millions of hours lost to driving could be spent w
No, doesnt answer my question. We will need to wait for the self drive unit is operational globally before improved safety claims can be verified. However the death toll for self drives has begun.
What is the total number of deaths [caused] by self driving vehicles?
I think it’s because they’re afraid that it may become the Antichrist, who, as I recall, should’ve been Al Gore... no, George W. Bush... no, Barack Obama... no, Hillary Clinton... no... umm... DONALD TRUMP.
"Why do some religious people fear artificial general intelligence?"
"Some People" You can't be serious. Another bunch of fluff BS that you plan to run on about for pages and pages and pages while making the same inane assertions you have made in your other 3 threads.
Pay attention now ----- "THERE ARE SOME PEOPLE WHO FEAR EVERYTHING." It's just that simple. The reasons do not matter.
@ Avant
Religious (and others) fear AI generally because of Horror/Fantasy/Science fiction popular since the 30's and reaching its zenith in the 60's and 70's
The Religious (unlike others) also fear AI and most medi biological research because it shows that man as a rational animal will be able to replicate the actions of their god in the act of creation, and particularly, if possible the creation of an intelligence with "free will". This will negate their claim to a special unique solitary being that can create and destroy, and bring crashing down the edifice of falsehoods surrounding their faiths,
Regarding the comments you. Avant, have made on other threads , you do seem obsessed with the future ramifications of AI. At the moment, and I say, at the moment, the concept of AI free from the constraints of its human programming is decades away if possible at all. What constraints AI will be forced to have by its creators, we do not know yet, because it is only a probability not a fact.
I predict we will all be long dead before anything approaching a demonstrable, working AI is even close to reality.
As Arakish as repeatedly pointed out the current state of the research has produced driverless autos, but it will be the 2080's before these are the norm, driverless trains and buses will be here sooner but they will still require human supervision, but not the one on one physical attention as today. ( I helped draw up the safety parameters for the driverless dump trucks on one of the biggest mines in the world) All examples of AI in use today are products of programming and constrained by the gigo principle. None of the algorithms you have quoted produce genuine free intelligence they are all constrained by the limitations of their creators...the programmers and hardware specialists.
Regarding your boast of being a PhD candidate. I am not surprised; the standard of ability and critical thinking required for any post graduate degree has fallen so much in the last twenty years I see no reason to doubt your claim.
(edit:spelling and punctuation)
1. People tend to be linear thinkers; scientists once predicted that the genome project would be completed in 700 years, after only 1% of it was completed in the first 7 years. Instead, it took only 7 more doublings in computing power, or 7 more years, an excellent example of exponential change [Wikipedia/Accelerating change].
~
2. It is odd that you cite Arakish, who has yet to provide any valid sources, only to reject the opportunity to cite his or her claims. Anyway, I guess laymen will prefer to cite laymen.
Albeit, there will reasonably be human level ai in 5 to 10 years, and that has many ramifications for several applications. This timeline is predicted by some of the smartest people in ai research, including Ray Kurzweil, who has a better track record than most people in his technological predictions, based on his "Law of Accelerating returns"
What metric are you using to make your predictions above? ( ͡° ͜ʖ ͡°)( ͡° ͜ʖ ͡°)
~
3. As to your disregard for my educational status, I've discovered that most persons here argue based on ignorance, and strangely, though they admit they are not experts in the field, they still offer claims of high confidence, that contrast evidence in ai research. Your opinion is thereafter noted, but as I've come to learn over the years, facts don't care about feelings, and the facts in ai research certainly are independent of your feelings/opinion.
Avant Brown: "People tend to be linear thinkers"
And some people just refuse to.
rmfr
@ Avant
1. Doubling of computer power: yes it is doubling of computer power every third year, in the cases of which I have practical knowledge, it means we can hook up more and more sensors to the ever more powerful computer. It is not AI, just more powerful than the original Atari with a much more sophisticated program. This just helps it meet it programmed outcomes. That is all.
2> It is not 'odd' that I cite Arakish. I would cite you if you made any sense at all.
Personally, I think, having real world experience, unlike your tyro self, in 5 - 10 years we will have 98% reliable driverless on call vehicles on our roads. It will be a further 20 years before they will be the majority on our roads barring legislation from governments. The paradigm shift in transport will be enormous. The transformation in industry will be as troublesome for society with most "road" vehicles (industrial) engineered for 20 year lives and most not compatible with aftermarket controllers. Tracked locomotives including trams are engineered for 50 year plus lives and are not suitable for aftermarket auto drivers.
Thats where real world practicalities and engineering interfere with children's dreams.
3. As I said, having met many recent PhD, the standards seem to have fallen into an abyss, where PhDs in Macrame and many other subjects are becoming as common as Halloween candy. They seem to be handed out after turning in some very average papers, and , I note are frequently offered for sale on that interweb thing. I suppose once we made the education system a dollar value thing, not a merit based achievement goal this was bound to happen.
1. Can you provide a citation wrt the scenario where you claim said instance of Doubling power above?
~
2. It is still demonstrably silly to cite somebody, who had yet to provide any citations to substantiate his arguments.
~
3. You are reasonably commiting a fallacy, by overgereneralizing the small pool of PhD related people you may have met, thereafter haphazzardly applying this to global standards. [See Wikipedia/Faulty Generalization]
Notably, I am doing PhD work regarding the combination of Quantum Field Theory with Artificial Neural Networks, so I think I am knowledgable in this domain, and probably more so that you, based on your expressions thus far.
~
4. In the future, please aim to cite valid resources. You may find that this may force you to rethink your expressions, although it may take a bit longer to respond, if you care about your expressions. Many people in academia tend to worry about their modes of expression, fearing risk of embarrasment, but you seem to be devoid of that concern.
*RE: "Regarding your boast of being a PhD candidate. I am not surprised the standard of ability required for any post graduate degree has fallen so much in the last twenty years I see no reason to doubt your claim."
I didn't think about this but, "How True." I would have been raked over the coals by my Mentors for such nonsense.
How will you earn the respect of your peers when a bunch of dweebs on an Atheist site can see through your nonsense? Do you even know how to construct a valid hypothesis? If this is what PhD programs have come to.... all I can say is "Wow! That's genuinely sad. Someone in your family has wasted a whole lot of money, or the mail order PhD company just found another sucker."
1. I didn't know PhD's were something to boast about. Somebody had made a remark regarding my education, and I had then responded appropriately. Notice that I had gone several pages, without posting anything about my academic status.
2. I have shown your efforts to be trivially demonstrably invalid.
Sadhu Sundar Selvaraj on the Jim Baker show? Seriously? *Cognostic's Shovel" OH MY GOD, THEY ARE BUILDING AN ARTIFICIAL BRAIN AND CALLING IT THE GODHEAD! AAARRRRRGGGGHHHHH! IT'S THE END OF THE WORLD. IT'S A BILLION TIMES SMARTER THAN A HUMAN. ARRRRGGGGHHHHH! ARMAGEDDON IS HERE..... JESUS WILL RETURN..... AI IS THE ANTICHRIST. AAARRRRRGGGGGGHHHHHHH!
Besides, got the LANL site and see the computational power they have. It is one of the greatest computer farms ever built. And they are only capable of having it do the same: run programs written by humans.
rmfr
Humans were "written" by nature/evolution.
Avant Brown: "Humans were "written" by nature/evolution."
Still showing off your lack of intelligence?
Check this out: Challenge
rmfr
Trying to move the goalposts? It is still being evasive.
Whether or not I am here to underline these truths, they are evident.
~
Notably, yes, most ai works today by solving tasks set by humans, but they do so from a blank-like starting state that the programmers encode. After that blank state, the algorithm learns to do the task without human intervention.
Crucially, the ai algorithms tend to learn billions of "parameters" or large collections or configurations of values between and inclusive of 0's and 1's, which no human programmer can sit down to imagine or write.
~
Aside from that, I also demonstrate that humans are not free from being programmed as well; humans are programmed by nature. The theory which describes this programming is called "evolution". So although no machines today have degree of autonomy that humans have, all things are ultimately programmed, as far as Science goes......
Good riddance to Avant Brown. It was fun the first day. Then exponentially, it became a raging hatred for the lack of mental capabilities in theists.
rmfr