Comments for AI Panic! http://aipanic.com This site is dedicated to research and unveil the perils, imminence and probabilities of a hostile takeover of the world through artificial intelligence. I will stay on the lookout for you and post articles, research papers and break-throughs of everything that could affect this danger. Thu, 11 Mar 2010 02:39:47 +0000 http://wordpress.org/?v=2.5.1 Comment on AI in Computer Games: A Threat? by Lynyrd Salita http://aipanic.com/ai-in-computer-games-a-threat/#comment-238 Lynyrd Salita Mon, 08 Mar 2010 04:24:13 +0000 http://aipanic.com/ai-in-computer-games-a-threat/#comment-238 true AI on games are made to lose their programs are too dumb to be compared to humans but think about this their too dumb to have reasoning that they only have one goal to kill the targets as possible as they could if applied to military purposes and robotics, they do not need reason, they only need instruction to follow and target to kill so is that harmless? you judge, in games, your character could have endless lives, but how about if your the character now and no weapons, no brute strength, no adv technology in recovery, no magic, etc, and robots that could repair, hack, reproduce, fearless and united would attack us, is there hope? true AI on games are made to lose their programs are too dumb to be compared to humans but think about this
their too dumb to have reasoning that they only have one goal
to kill the targets as possible as they could
if applied to military purposes and robotics, they do not need reason, they only need instruction to follow and target to kill
so is that harmless?
you judge, in games, your character could have endless lives, but how about if your the character now and no weapons, no brute strength, no adv technology in recovery, no magic, etc, and robots that could repair, hack, reproduce, fearless and united would attack us, is there hope?

]]> Comment on State of Google AI: A Long Way To Go by Randy http://aipanic.com/state-of-google-ai-a-long-way-to-go/#comment-237 Randy Tue, 02 Mar 2010 04:24:36 +0000 http://aipanic.com/?p=55#comment-237 I am not quite sure why this is such a problem...? It makes sense to me at least. Type in a larger number in a smaller measurement of a larger one it relates to. Such as 13 inches in a foot... or 6000 feet in a mile... or 20 months a year. You basically are asking the Search Engine what is X amount over 100% in this subject. I hope I am not incorrect on such but mathematically this is correct and logically it makes sense as well. I am not quite sure why this is such a problem…? It makes sense to me at least. Type in a larger number in a smaller measurement of a larger one it relates to. Such as 13 inches in a foot… or 6000 feet in a mile… or 20 months a year. You basically are asking the Search Engine what is X amount over 100% in this subject. I hope I am not incorrect on such but mathematically this is correct and logically it makes sense as well.

]]> Comment on Japanese Nanobot Brain To Control The Goo by Daniel Hazelton Waters http://aipanic.com/japanese-nanobot-brain-to-control-the-goo/#comment-233 Daniel Hazelton Waters Tue, 09 Feb 2010 01:00:48 +0000 http://aipanic.com/japanese-nanobot-brain-to-control-the-goo/#comment-233 Long before nanobot brain enhancers. There will be computers that use memristors and then quantum computers. These computers will transform the electronics revolution. There will be science fiction like qualities to come. Long before nanobot brain enhancers. There will be computers that use memristors and then quantum computers. These computers will transform the electronics revolution. There will be science fiction like qualities to come.

]]> Comment on Researchers Create Smartest AI, Say Adult Level Intelligence Very Far Away by Franklin http://aipanic.com/researchers-create-smartest-ai-say-adult-level-intelligence-very-far-away/#comment-230 Franklin Wed, 14 Oct 2009 19:20:26 +0000 http://aipanic.com/researchers-create-smartest-ai-say-adult-level-intelligence-very-far-away/#comment-230 I don't get it. At one point, the article says that the system has the intelligence of a two or three year old child, yet the article says towards its beginning that the system has the intelligence of a puppy. But a two or three year old is a lot smarter than a puppy. Which one is it? I don’t get it. At one point, the article says that the system has the intelligence of a two or three year old child, yet the article says towards its beginning that the system has the intelligence of a puppy. But a two or three year old is a lot smarter than a puppy. Which one is it?

]]> Comment on The Humans Are Dead by Storm http://aipanic.com/the-humans-are-dead/#comment-224 Storm Mon, 31 Aug 2009 22:42:12 +0000 http://aipanic.com/the-humans-are-dead/#comment-224 I love the "binary solo" part. I love the “binary solo” part.

]]> Comment on State of Google AI: A Long Way To Go by Eric L http://aipanic.com/state-of-google-ai-a-long-way-to-go/#comment-222 Eric L Sun, 23 Aug 2009 03:35:14 +0000 http://aipanic.com/?p=55#comment-222 Your post seem to imply that Google is applying the best minds of the area on building strong artificial intelligence in the Google search query box. This is not the case. The brightest minds are trying to make the most reliable and consistent output, not the most sentient. Google search query bar is not going to be the origin of sky net, I agree with you 100% there. But I think to fully make the argument your trying to make, you have to show a little effort in finding programs that exhibit some of sky net's capabilities. For example: Botnets and stealth malware capable of infecting multiple platforms through an array of tactics implementing learning algorithms to gather new resistance methods and attack vectors from the discoveries and perils of bots and malware on other machines. Essentially a self-evolving program. This would be where skynet will build his first killer app (pun intended). Your post seem to imply that Google is applying the best minds of the area on building strong artificial intelligence in the Google search query box. This is not the case. The brightest minds are trying to make the most reliable and consistent output, not the most sentient.

Google search query bar is not going to be the origin of sky net, I agree with you 100% there. But I think to fully make the argument your trying to make, you have to show a little effort in finding programs that exhibit some of sky net’s capabilities.

For example:

Botnets and stealth malware capable of infecting multiple platforms through an array of tactics implementing learning algorithms to gather new resistance methods and attack vectors from the discoveries and perils of bots and malware on other machines. Essentially a self-evolving program. This would be where skynet will build his first killer app (pun intended).

]]> Comment on Researchers Create Smartest AI, Say Adult Level Intelligence Very Far Away by God http://aipanic.com/researchers-create-smartest-ai-say-adult-level-intelligence-very-far-away/#comment-219 God Sat, 13 Jun 2009 22:18:50 +0000 http://aipanic.com/researchers-create-smartest-ai-say-adult-level-intelligence-very-far-away/#comment-219 ME. I am the first TRUE A.I ME. I am the first TRUE A.I

]]> Comment on State of Google AI: A Long Way To Go by Pinner http://aipanic.com/state-of-google-ai-a-long-way-to-go/#comment-218 Pinner Fri, 29 May 2009 18:15:19 +0000 http://aipanic.com/?p=55#comment-218 This result is returned on Google's UK site, but not on it's US site. Clearly geography is very "important". This result is returned on Google’s UK site, but not on it’s US site. Clearly geography is very “important”.

]]> Comment on Anti-Landmine Group Campaigns Against Autonomous War Robots, Wired Thinks That’s Stupid by Unit 01 http://aipanic.com/anti-landmine-group-campaigns-against-autonomous-war-robots-wired-thinks-thats-stupid/#comment-215 Unit 01 Sat, 11 Apr 2009 11:25:54 +0000 http://aipanic.com/anti-landmine-group-campaigns-against-autonomous-war-robots-wired-thinks-thats-stupid/#comment-215 While I agree that reckless development and deployment of autonomous armed devices might pose a threat, I still think that, if properly implemented (yup, I know that is one hell of a caveat) these "fully automated armed military systems" will lead to more precise application of military force and less civilian bloodshed and suffering. A war waged through properly implemented autonomous "war-bots" will be more humane towards noncombatants. Why? Because: 1) A fully automated armed military system is not bound to have self-preservation as strong as a human combatant. While human soldiers are likely to use lethal force in response to a slight hint of unfriendly behavior (even if ordered not to do so) due to inherent human self-preservation instincts and combat stress, a "war-bot" which lacks such human instincts will strictly follow the "innocent until proven guilty" maxim if ordered to do so, thus preferring to err on the side of misdetecting a combatant over accidentally killing the proverbial "girl who tries to offer war-bot an ice-cream". A properly implemented fully automated armed military system will not suffer "morale loss" over the fact that following such orders increases the threat to its existence. Also, a war-bot following such "civ-friendly" orders will not incur the horrible costs in regards to morale at home, as there are no friends or relatives to mourn a war-bot. 2) War-bots are EXTREMELY unlikely to develop truly sadistic inclinations (unless programmed to do so) 3) Properly implemented war-bots will not develop "war-time nationalism", thus will not be hostile to all people belonging to the ethnicity common among enemy combatants 4) A properly implemented fully automated armed military system will not demonstrate "personality traits" (not a term that suits a robot well, but I hope you forgive me using it to describe certain behavioral patterns common to humans) that are likely to instill hatred and hostility in civilian populations in the warzone and at home (robots will not throw puppies) 5) A properly implemented fully automated armed military system will not demonstrate negative "personality traits" usually arising due to war-time stress and overall criminologenic environment of war (robots will not rape women "because they can)... Unless programmed to do so ;) 6) A properly implemented fully automated armed military system will demonstrate "personality traits" that will instill trust, friendliness and generic desire to support the "war-bots" in allies and non-combatants. 7) Once a "skill" (like "detect suspect suicide bombers from gait, heartbeat and other behavioral and remotely-detected physiological properties") is implemented in AI, replicating it to other machines is cheap and easy. Replicating skills in humans is expensive, hard, and not very reliable. Off course, "civ-friendly" war-bots will suffer additional losses due to guerrillas taking advantage of their "civ-friendliness", but the pace of technological progress will likely make the costs of each individual autonomous war machine comparable to that of training and equipping a human. Also, the sheer superiority an advanced war-bot will have in terms of armor, firepower and sensor capabilities will make it harder for human enemy to take advantage of. It is also possible that AI will have additional analytical capability allowing them to deduce hostile intent from evidence that would not be sufficient for a human mind to reach same conclusions (it does not mean the AI will be "smarter than human", it will be merely "better at spotting IEDs") All-in-all, I think it is possible that, thanks to AI research and advanced robotics., we will wage safer, more human-friendly wars in the future. While I agree that reckless development and deployment of autonomous armed devices might pose a threat, I still think that, if properly implemented (yup, I know that is one hell of a caveat) these “fully automated armed military systems” will lead to more precise application of military force and less civilian bloodshed and suffering.

A war waged through properly implemented autonomous “war-bots” will be more humane towards noncombatants.

Why?

Because:
1) A fully automated armed military system is not bound to have self-preservation as strong as a human combatant.
While human soldiers are likely to use lethal force in response to a slight hint of unfriendly behavior (even if ordered not to do so) due to inherent human self-preservation instincts and combat stress, a “war-bot” which lacks such human instincts will strictly follow the “innocent until proven guilty” maxim if ordered to do so, thus preferring to err on the side of misdetecting a combatant over accidentally killing the proverbial “girl who tries to offer war-bot an ice-cream”.

A properly implemented fully automated armed military system will not suffer “morale loss” over the fact that following such orders increases the threat to its existence.

Also, a war-bot following such “civ-friendly” orders will not incur the horrible costs in regards to morale at home, as there are no friends or relatives to mourn a war-bot.

2) War-bots are EXTREMELY unlikely to develop truly sadistic inclinations (unless programmed to do so)

3) Properly implemented war-bots will not develop “war-time nationalism”, thus will not be hostile to all people belonging to the ethnicity common among enemy combatants

4) A properly implemented fully automated armed military system will not demonstrate “personality traits” (not a term that suits a robot well, but I hope you forgive me using it to describe certain behavioral patterns common to humans) that are likely to instill hatred and hostility in civilian populations in the warzone and at home (robots will not throw puppies)

5) A properly implemented fully automated armed military system will not demonstrate negative “personality traits” usually arising due to war-time stress and overall criminologenic environment of war (robots will not rape women “because they can)… Unless programmed to do so ;)

6) A properly implemented fully automated armed military system will demonstrate “personality traits” that will instill trust, friendliness and generic desire to support the “war-bots” in allies and non-combatants.

7) Once a “skill” (like “detect suspect suicide bombers from gait, heartbeat and other behavioral and remotely-detected physiological properties”) is implemented in AI, replicating it to other machines is cheap and easy. Replicating skills in humans is expensive, hard, and not very reliable.

Off course, “civ-friendly” war-bots will suffer additional losses due to guerrillas taking advantage of their “civ-friendliness”, but the pace of technological progress will likely make the costs of each individual autonomous war machine comparable to that of training and equipping a human. Also, the sheer superiority an advanced war-bot will have in terms of armor, firepower and sensor capabilities will make it harder for human enemy to take advantage of.

It is also possible that AI will have additional analytical capability allowing them to deduce hostile intent from evidence that would not be sufficient for a human mind to reach same conclusions (it does not mean the AI will be “smarter than human”, it will be merely “better at spotting IEDs”)

All-in-all, I think it is possible that, thanks to AI research and advanced robotics., we will wage safer, more human-friendly wars in the future.

]]> Comment on Japanese Nanobot Brain To Control The Goo by sir.patrick http://aipanic.com/japanese-nanobot-brain-to-control-the-goo/#comment-214 sir.patrick Sun, 08 Mar 2009 07:24:22 +0000 http://aipanic.com/japanese-nanobot-brain-to-control-the-goo/#comment-214 Mankind is not ready for naobots interacting with our neurons. yes we may have no more use for cell phones or compurters beccause we will be connected to the internet thur our nonobot neurons and we will have virtural reality when ever we want it.Some geek will be trying to controll the nonobot neurons in my brain and controll me and my finances. The military will add nonobots to the water and we will all be under there control.I hope you have a good password to the nonobots in your brain. Those without nanobots injeceted into there neurons well be the underclass. Replicators will be the next version of microwaves in our kitchen. Put a piece of coal and a little water and just ask the replicator what you want it tom make. Like that Woody Allen move and Star treck with orgasmatrans,replicators and nonobots to infest your enemeys host to eliminate them like they did with the liquid space aliens. It is not if but when and where nonobots well be in your body. God help us when that happens. Mankind is not ready for naobots interacting with our neurons. yes we may have no more use for cell phones or compurters beccause we will be connected to the internet thur our nonobot neurons and we will have virtural reality when ever we want it.Some geek will be trying to controll the nonobot neurons in my brain and controll me and my finances. The military will add nonobots to the water and we will all be under there control.I hope you have a good password to the nonobots in your brain. Those without nanobots injeceted into there neurons well be the underclass. Replicators will be the next version of microwaves in our kitchen. Put a piece of coal and a little water and just ask the replicator what you want it tom make. Like that Woody Allen move and Star treck with orgasmatrans,replicators and nonobots to infest your enemeys host to eliminate them like they did with the liquid space aliens. It is not if but when and where nonobots well be in your body. God help us when that happens.

]]>