What If Gamers Taught AI / Jade Helm Program Psychology?

a reply to: FissionSurplus

Destiny has a "Nightfall" mode. No respawns.

You have to run the mission with your other two teammates without all three dying. Sure, you can resurrect your fallen comrades, but they have timers on them so you can't revive them instantly. If all three of you die, you have to start all over. I can't tell you how stressful it is to be the last team member alive, trying to formulate a way to revive your friends when under fire.

It certainly makes for more tactical play-style. You really don't go rushing into an area before evaluating the threat level.

a reply to: FissionSurplus

Honestly, I think the A.I. you are proposing would just grief humanity into giving up.

In all seriousness tho, I have yet to find a single computer game (especially one as mundane as our online games have become) that can actually compete at all with my imagination. Nine times out of ten I am disappointed and go back to reading or writing. Computer games at the moment can only really interact with people in a basic level, there is very little creativity involved in going through a chain of action sequences plus movie/dialogue.

That being said, there is some great scifi and Japanese Anime that touches on all this from a video game perspective. Spielberg is apparently making Ready Player One.

a reply to: MrSpad

American army was never officially used to train solders but ARMA was.
American army has been used by the army in virtual reality type setup just as arma was/is but not "officially

I take it your not a gamer? I have been competitively playing video games for at least 8 years.

For the sake of discussion, lets say that the game Battlefield 4 was made for this. Battlefield 4 is a large FPS game with jets, helicopters, tanks and a wide range of weaponry. Lets not forget about the amazing large life like maps. Not only can you play as a solider but you can play as a commander as well. You can guide your team and tell them what to do.


Video games are limited compared to reality, you cant load your guns magazine faster then the game allows, you cant run faster or walk faster, compared to reality the amount of control you have over a character is very limited BUT that does not mean a good amount of information cannot be minded from these games. While the physics and animations limit the game a human being is still controlling it. A AI could learn a WHOLE lot about war just by watching a chess games. Video games provide far more information then a simple chess game especially when it comes to accurate statistics considering how many people game. FPS games would provide a grate deal of information but even games like world of war craft, Dota 2 or civilization have a grate potential.

I think that if it is possible to this with our technology it would be logical/highly probable that we are. It would seem it is possible considering the amount of information we are all ready mining. If not by a AI by some physiological science organization.

It could be used for combat but it has other uses as well. Take for example a game like sims, just imagine all the information a AI system could gain from a game like sims. A AI does not have a body so it cant venture out and interact with the world like we do, but in a game like SIMS gives it has a opportunity to do so. Ok maybe its not physical reality but psychologically speaking its VERY close.

a reply to: Thorneblood

My family never plays online. We love the system link games, especially the forge option in Halo! My husband likens it to Lego fun in building our own maps and rules of play. We now have hundreds of customized maps that our friends love to share via USB. Our xboxes will never be connected to the Internet gaming. Some of our friends have to share our straight boxes since the updates will totally mess up their direct syncing abilities! Ugh!

As far as, respawning, it is true that one may play more recklessly than they do when they cannot. Honestly, though, I think if you are going to fight a fanatic that believes there is life after death...there is no difference! So if they are computing this data into AI data base, it would actually be quite a wise decision.

Thanks for sharing your thoughts and the added info.

JadeHelm talk, probably due being worm out with all the talk and hype before hand, has fell silent since it's begun this week. The going concern fears are Marshall law. This helps to update those unaware that there's potentially more and long run concerns than suspected. The A.I. might really be the next step in defense of the US. But A.I. comes with warnings even by tops researchers.

Also, I agree that the monitoring implanted into these gaming systems is more than just marketing. Some of the data collections could very well be used in tests to establish a sort of "pre-crime" detection system when creating a virtual world comprised of selected avatars-of certain demographics for example, to determine outcomes.

Some might say I fell victim to a transhumanist ideals when explaining interest in virtual video game worlds. And with, such as, it's conditioning for the post humanity future. If that's so, doubt us normal citizens will be included in immortality(well, and many probably wouldn't) through your uploaded "self"(as science explores to prove consciousness) into a virtual avatar. The first step in the process, is course robotics, to harbor the consciousnesses.

originally posted by: JamesCarrot
a reply to: MrSpad

American army was never officially used to train solders but ARMA was.
American army has been used by the army in virtual reality type setup just as arma was/is but not "officially

I take it your not a gamer? I have been competitively playing video games for at least 8 years.

For the sake of discussion, lets say that the game Battlefield 4 was made for this. Battlefield 4 is a large FPS game with jets, helicopters, tanks and a wide range of weaponry. Lets not forget about the amazing large life like maps. Not only can you play as a solider but you can play as a commander as well. You can guide your team and tell them what to do.


Video games are limited compared to reality, you cant load your guns magazine faster then the game allows, you cant run faster or walk faster, compared to reality the amount of control you have over a character is very limited BUT that does not mean a good amount of information cannot be minded from these games. While the physics and animations limit the game a human being is still controlling it. A AI could learn a WHOLE lot about war just by watching a chess games. Video games provide far more information then a simple chess game especially when it comes to accurate statistics considering how many people game. FPS games would provide a grate deal of information but even games like world of war craft, Dota 2 or civilization have a grate potential.

I think that if it is possible to this with our technology it would be logical/highly probable that we are. It would seem it is possible considering the amount of information we are all ready mining. If not by a AI by some physiological science organization.

It could be used for combat but it has other uses as well. Take for example a game like sims, just imagine all the information a AI system could gain from a game like sims. A AI does not have a body so it cant venture out and interact with the world like we do, but in a game like SIMS gives it has a opportunity to do so. Ok maybe its not physical reality but psychologically speaking its VERY close.

I play games and I have been in combat and they are nothing remotely alike. Games are games, you can teach some basic tactics with things like chess or games like Hearts of Iron but, you can learn nothing from first person shooters.

a reply to: MrSpad
Psychology and social behaviors, though. Tendencies of similar personality types, etc. Wouldn't you think?

a reply to: The GUT

There's too many trolls in gaming for any information to be even remotely viable. Especially in shooters.

a reply to: Flesh699

There are equally as many who take it seriously.



Interesting thoughts, FS, and certainly something to scratch the cranium over.

Taking a further step back, going with the idea that games can teach an AI and tie it in to the idea that we are living in a simulation and you have yet another layer of removal for a teaching/learning relationship.

Maybe we're in a Thirteenth Floor scenario and just haven't quite yet found the portal to the next nigher realm...

a reply to: FissionSurplus

There's a big problem with your premise. The object of just about every game is dominance, the go to method of dominance in strategy is resource attrition multiplied by production. If your opposition can produce 10 units/second but needs to spend 11 units/second to keep up in combat and has 500 units in reserve the battle will be lost in 500 seconds. If you as the aggressor only produce 8 units/second but spend 10/second to maintain the battle but you have 1300 units in reserve, you can maintain for 650 seconds. Therefore despite lower production and worse efficiency you can still win because the game ends after you achieve your objective in 500 seconds.

Real life does not work this way, after you achieve an objective you still need to maintain control over an area, diplomatic changes, trade opportunities, and so on. There are a lot of factors that come into play beyond merely winning the game because the game never ends, the rules simply change from decade to decade.

Another example is the rush which is a hallmark of strategy games. It involves sacrificing all mid and late game potential to send a small force of poorly trained and ineffective units after an enemy before they have any defenses set up. It is costly in bodies and extremely costly in economic development but makes for one of the best RTS strategies available, so much so that game developers put roadblocks to prevent it in todays games.

The last example is attacking resources. In a strategy game, the best way to win is to kill your opponents resource collectors which would be analogous in real life to targeting civilians. In an RTS or turn based game this stops a persons economy while yours pulls ahead, which inevitably leads to better units. Worker units cannot convert into military units in most strategy games. In real life the exact opposite is true, if you target the workers other nations will join the fight against you on humanitarian grounds and civilians routinely convert into military units by joining the army or wearing a suicide vest.

Any AI that attempted to win a war based on video game tactics would lose very quickly.

originally posted by: JamesCarrotVideo games provide far more information then a simple chess game especially when it comes to accurate statistics considering how many people game. FPS games would provide a grate deal of information but even games like world of war craft, Dota 2 or civilization have a grate potential.

Competitively playing games? I've played them professionally, though I don't do that anymore. Lets speak of Civilization 4 (we can pretend 5 doesn't exist), I'm actually credited on the games expansions because of AI input I provided after having been one of the first people to beat Deity, and writing the guides to let others do so.

You have Warriors as your only military unit and your opponent has Archers. It takes roughly 3 Warriors (90 hammers on Marathon) to kill a single Archer (50 hammers), while the AI also has a 33% production bonus, and starts with 2 free Archers. This means that to take a nearby capital you need to spend roughly 360 hammers before your opponent can spend 66. It is impossible to do this without expending all possible resources (chopping lots of forests). However there's a flaw in the AI. If the AI sees you have a lone unit that it has great combat odds against, it will move a unit out of the city to snipe it as a free kill. Because of this you can place a stack of Warriors 2 tiles out from their city, and then put a single Warrior 1 tile out from their city. An archer will move out to kill it and be heavily damaged because it's on offense rather than defending in a city (resulting in the loss of several combat modifiers). Then you can move a single Warrior to finish off the archer. The AI will respond by seeing a damaged archer outside the city gates and attack it. Using this strategy you can trade 1 Warrior (30 hammers) for the AI's Archer (33 hammers after production bonus). Eventually doing so will exhaust their defenses and allow you to take the city at 1/6 the cost of simply assaulting it.

This strategy is completely infeasible in real life. Any AI that recommended doing so would be instantly dismissed by any moderately capable tactician.

a reply to: FissionSurplus

That conspiracy theorist/survivalist Dr. Bill Deagle used to say that his 'insiders' told him, like over 10 years ago, that the NSA and others had input the entire system of World of Warcraft and other games such as that to monitor the players - possibly for future use in virtual war situations. I am not sure. He is also scared the crap out of people about flu epidemics... and just so happened to have all the supplements and cures to sell! Like Steve Quayle... But I think a lot of what he said was probably based on truth.

originally posted by: corsair00
That conspiracy theorist/survivalist Dr. Bill Deagle used to say that his 'insiders' told him, like over 10 years ago, that the NSA and others had input the entire system of World of Warcraft and other games such as that to monitor the players - possibly for future use in virtual war situations. I am not sure. He is also scared the crap out of people about flu epidemics... and just so happened to have all the supplements and cures to sell! Like Steve Quayle... But I think a lot of what he said was probably based on truth.

I've worked on MMO's in addition to being a very high end player in them. There is no secret code to monitor players, and even if there was the data you would get for it is pretty useless. Raid and group ability in MMO's isn't about strategy at all, there are many valid strategies to do just about everything. What actually matters in them is personnel management and ensuring people can perform whatever task the strategy calls for at a given time. This is why the absolutely most difficult thing you can create as an MMO developer is a large raid event that requires each person to act autonomously to individually changing circumstances.

No contest, IMHO.

AI and RPG games aren't even a foot in the door when it comes to real combat scenarios.

Even if they perfect that "no fear" drug, there will always be measures that can't be accounted for beforehand.

It's a far stretch.

no, just no. games are retarded compared to the streets, let alone a combat zone.

they are boring, boring, and.... boooorrring. i'd MUCH rather get lost in my mind than let muscle memory do it's thing and not be building a bod in the process. ridiculous waste of time

a reply to: Aazadan



You GUYS!!!!

Its not about the games limitations or reality vs games

Its about how the player (the human) plays the game. The psychology behind there actions, choices, reactions ect...

Aazadan you know a lot about civ4, that alone can tell the AI allot about your personality.
The moves you make in order to win the game, your reactions to the enemy, the choices you make ect.... will give a little bit of information about your personality especially if you make the same choices over and over.

Hows does jumping off high heights in Halo to dying from two bullets, instead of one to the head like COD online, would be considered vital data?

A simulator only as good as to how realistic it is, and we don't have games that could effect our sense of touch yet ether.

a reply to: FissionSurplus
Maybe! Maybe not!

Would make a cool story either way.

Though we can use some more smarter bots, the ones we got now kind of suck.

What we need is the one bot to rule them all. Oh ya I can see it now. Forged in mount doom by the anvil of crom, created and programed in basement of oblique contexts and cat videos, patented at Microsoft, but made in china, then packaged in Taiwan, then shipped and labeled in Mexico, and then finally shipped to the US for assembly. After which?

Oh ya! Were totally doomed. Once again we will get a shoddy product.

If that were the case, I'd want my money back - and paid for training instead. lol

originally posted by: JamesCarrot
a reply to: Aazadan

You GUYS!!!!

Its not about the games limitations or reality vs games

Its about how the player (the human) plays the game. The psychology behind there actions, choices, reactions ect...

Aazadan you know a lot about civ4, that alone can tell the AI allot about your personality.
The moves you make in order to win the game, your reactions to the enemy, the choices you make ect.... will give a little bit of information about your personality especially if you make the same choices over and over.

I might know a lot about 4, I play it a lot to this day because I firmly believe it's the best TBS game ever built but I'm much worse at it than I used to be.

Anyways, you may be interested to know that both chess and poker are well studied in how decisions relate to military matters. From a few games of poker and chess you can get a very good feel for a generals command ability. I don't know if this is still practiced today but it was quite popular in the past.

I don't play much poker, but I do play a lot of Magic the Gathering to some reasonable success (winning some state championships and such, but I'm not in a living situation where I can do the extensive travel the game needs to go bigger than that, and I'm probably not good enough to either). I play the game in both paper and online and there are a lot of small things that go on beyond the game actions. For example, if my opponent draws a card, shuffles his hand, and then looks at two cards, I can deduce what those 2 cards are to a short list of 3-5 each and depending on the cards my mana and deck type are representing I can deduce if the first of the two cards is the weaker test card or the stronger one he actually wants. If for example we're doing a cube draft and my opponent leads on turn 2 into my open mana with a Pack Rat with no hesitation, it's likely my opponent wants to go all in on that plan. If they think it over a bit (including thinking it over on the previous turn) it's more like that the Pack Rat is must answer bait for something even more likely to win the game. In that case I would want to let the Pack Rat live and try to beat it, rather than kill it. This all involves non verbal signals based on the opponents actions.

This is the problem with games in general. We can teach all the rules, and computers are quite good at following the rules but human responses aren't rational. Going against an AI in a brand new game, you'll search for ways to repeat behavior, and then abuse it. Going against humans is something else entirely because humans aren't logical, and unlike the computer they are not compelled to play the game to it's conclusion. Going back to civ 4, for a long time when multiplayer was big for that game you would have 16 person games, all 16 people would rush for the pyramids, whoever got it would have a very powerful world wonder but 10 others would drop out of the game. If you were studying that behavior it involves people taking a lot of high risk/high reward actions and then walking away if it doesn't work.

If an AI were studying my personality in games, assuming it could write an indepth enough report what it would find is that before I take any action I think many moves ahead, I stick to turn based games so this is easy to do. Before I place my first city in Civ I'm thinking about what nearby tiles are city 2, and how they coexist, and then I'm thinking about what area is most defended. In this case I think about 400 turns out. Before I play my turn in Magic I look at all the ways the game can go based on known/derived/assumed information and figure out how to counter the lines I can't beat (and there's some that I simply can't win, such is the nature of the game). This is similar in other strategy games too, I think about how I want to win, and then go down the lines to make that happen.

What any report on my personality would find is that I'm extremely inconsistent and this is true of most people. Given the same information x% are going to do one thing with it while x% are going to do something else. The same data creates different responses, which will each go on to create a different outcome. What this means is that you're going to find that situational circumstances have little bearing on a persons actions in the real world. People act on a macro level in a few different categories.