What Would You Like to See from Game AI?

jtogel asks: "As someone working in new (bio-inspired) AI research with an eye to applications in games, but within an academic setting, I often hear that game developers are not incorporating cutting-edge academic AI into their projects because it's too "risky" (they can't really predict how gamers would react), and because they don't see the point in it. As a gamer, and as someone who cares what gamers think, I am often surprised by the sorry state of current commercial game AI - it has hardly moved since the 1980s. However, maybe the problem is that no-one really knows what we want from game AI. Academics keep coming up with innovative AI technologies, but what we should we use it for? What do you think? What sort of intelligent behavior would you like to see in games, but don't at present? Which are the most obvious intelligence deficiencies of current NPCs that need to be fixed?"

Comments

Turing

[eldavojohn]

For the purposes of this discussion, I will target the two extremes of games. We have game type I in which the game space is very much finite and complete search trees are exploitable and can easily be enumerated.

Let us take the simple game of tic-tac-toe (three by three) as our game space. Not too hard. Tic-tac-toe, checkers, cognac, are popular games that fall in this category. And when you use AI to its full advantage, the computer rarely loses.

As we move towards type II games, we see games like chess or go where the game space is too large to search but still 'algorithms' (or heuristics) can be defined that prune the tree space or match/strive patterns out of raw moves. Look up tables are also useful but not really "AI" in the strictest of sense.

A full fledged type II game would be something like Warcraft or a complex computer game in which the internal engine of the playing board (server in this case) has an innumerable amount of states that it can exist in.

The NPCs and AI in games like Doom or even Halo are still fairly simple. They rely on heuristics and Euclidean distances (conceptually within the game) to overcome their opponents. If they are faced with obstacles, they deviate from their path.

Now you ask me what I would like to see. I have very finite desires and I will list them here:

• AI that learns. I am so sick of AI not keeping a log of encounters. There are many learning algorithms out there and none of them apply to full emersion games (like SWG or WoW) and its because there are too many variables--too many parameters to build the small dimensional vector fields of the Win or Learn Fast (WoLF) algorithm. This isn't the only kind of learning though, what about statistical analysis of prior opponents? "I'm a level 54 mage encountering a level 52 warrior, my odds of survival are x% based on past encounters with warriors two levels below me..." Yes, it's pre-canned statistical analysis but things like this could seriously act as a good heuristic for many games.
• If you reduce the game space or find some really good heuristic (as in the above), dumb it down. Meet the user's pleasure level. It should be a challenge but not such a challenge that victory is unattainable.
• Use your imagination. That's all I'm going to say about that one. There is no cookbook for heuristics ... learn to throw things at the user.
• Introduce long term "bounty hunter" style AI. I would enjoy an AI that sleeps and is constrained like I am. Introduce simple work parsers to our enemies. The closer you can come to interacting with me on the level of a real human, the more I'm going to like it. I know there are robots out there that understand basic human speech, why can't there be games where we receive phone calls or e-mails from assassins in real life? I know this sounds ridiculous, but make me afraid of my foe outside the game. Get in my head.
• Introduce random variables. I don't care if you have to build in mechanisms to your game that are illogical so that the AI sometimes goofs up. All humans are fallible at a certain point and if you have differing levels of predictability in your AI, the user will love it.
  
  

In the end, my opinion is that the largest deficiency for game AI is that it fails the Turing Test.

Horribly.

I can easily tell that I'm not playing another human.

Re:Turing

[joshdick]

"Introduce random variables. I don't care if you have to build in mechanisms to your game that are illogical so that the AI sometimes goofs up. All humans are fallible at a certain point and if you have differing levels of predictability in your AI, the user will love it."

Sure, NPCs should goof up now and again, but most of the time, they should at least be rational; they should always take steps not to die.

That means:
* No NPC should stand by idly as I charge toward it with weapons drawn.
* If I swiftly and easily kill all N-1 enemies in a group, the Nth one ought to get a clue and run away with its tail between its legs, beg for mercy or at least try a different strategy.
* No NPC should forget about me simply because I walk out of its view.
* NPCs should use teamwork.

Also, there should be variability among NPCs, even of the same type or class. For example, in "Halo," all Elites are the same, so once you figure out how to beat one, you'll be able to beat 'em all.

Maybe games have to advance for AI to advance

[snookumz]

The developers of Red Steel for the new nintendo wii said that they upped the AI players combat skills, because the new controller allowed faster and more accurate aiming than a traditional controller. They're point was that game AI is usually braindead to make it reasonable to the players who are slightly hindered by an unnatural interface. It's common in video games for a character to stand without any cover at the end of a long hallway. This isn't because the programmers couldn't program the AI to look for cover. Game characters aren't often built to be tactical, because they'd creme the gamer. That's my take.

Look to FramSticks for an example

[maynard]

FramSticks

Here is a great example of Artificial Life that generates truly unexpected behavior during runs. Perhaps this type of Alife simulation could be an inspiration for a new generation of game AI. Do not program a location and series of behavior patterns, instead make a population of AIs based on a variety of physical forms. Each form will have a limited set of possible movements within a the simulated world. It will need inputs in order to determine friend and foe, perhaps something similar to limited vision and hearing. It will need survival as a baseline goal. They will also need "food" of some sort. Perhaps the player or other, different, AIs can represent a food source. Give it some form of sex in order to reproduce its learned behavior through some genetic mechanism.

I suspect a hardware physics chip would help tremendously. But what I've seen of FramSticks was pretty damn cool. I have no idea how well it could be incorporated into AI gaming though. So this is just one of those: *shrug* hey, what about this? type of posts from someone ignorant and totally out of the field.

I don't know what I want to see

[Pluvius]

But I'll tell you what I don't want to see: Dynamic AI. Or at least the kind of dynamic AI we see now-a-days. I'm talking about those racing games where you're beating the hell out of the computer driver but in the last lap or so he ends up becoming completely inhuman (perhaps even "cheating") and beating you or coming close to it. Stop that shit, it's not any fun.

Rob

I want AI to be obvious

[silvermorph]

AI isn't currently fun. It's used mostly as a difficulty slider. And when it's doing its most impressive things, I don't see it happening, so I don't care. So I want two things from AI: I want to know what it's doing, and I want to be the cause of what it's doing.

1) If I've got a bunch of bots on my team, and I rush out like a madman, I want them to be completely suicidal as well, or maybe cover me. And I want them to say "covering you" as they do it, so I don't go sit behind a rock as soon as I hear gunfire.
2) If I'm sneaking up on some AI enemy, I want the AI to recognize that I'm sneaking up on it, and then LET ME SNEAK UP ON IT. Maybe it notices that I'm right next to something I can duck behind, so it shuffles its feet and slowly looks around back towards me, giving me just enough time to respond.
3) If I'm in a deathmatch, I want clever, flawed, fair bots. Bots that know good spots to fight from, but get up and move after a kill or two. Bots that aim poorly when I get the jump on them they just picked up their mouse in panic.
4) If I'm fighting a squad of enemies, I want to overhear their communication somehow so that I don't think they're looking through walls waiting for me. And I want them to react slowly to what I'm doing, not instantaneously.

I want my AI to be fun, and I want it to be pretty obvious it's being fun.

Artificial worlds need to feel self-sustaining

[Dr. Spork]

I know first hand that it's possible to write excellent combat AI as an ordinary (though long) nested list of conditionals. I'd call that hard-coded intelligence. This is the sort of thing I'm pretty good at (for example, the smart Irenicus in the Ascension mod for BG2, who's deadly without cheating). I have a feeling that most gamers are satisfied by this sort of AI if it's really done well, so that smart NPC's really fight smart.

But real AI has to involve some sort of learning, which is to say, letting game events "write" your behavior script. When would this be useful? The best example I can think of: Entirely stable environments that are "alive". Current games give you staged non-equilibrium situations that get triggered when your PC enters the scene. This sort of thing is just very obvious and unsatisfying if the goal is immersion. What good AI might do is this: before the game is released, the various separate settings might be populated with a bunch of artificial-life characters with specific motivations, needs, preferences, etc. (Maybe like the Sims, except more complex psychologically.) Then the game authors would let this initial system reach an equilibrium their big server. If they don't like the equilibrium that was produced, they tweak the initial AI and try again. Eventually, this will produce in a "natural" way something like a small, functioning village. When a PC enters the village, it will have been in an equilibrium which the actions of the PC will disrupt, almost certainly in unpredictable ways. That is how you give the player a true sense of freedom, like their actions really matter. Somebody like me might wonder: What would happen if I steathily killed the village miller, or gave him a gigantic horde of treasure, etc? That sort of scenario is impossible to play out in current games. And that sucks.

Now granted, writing Artificial Life that reaches an equilibrium similar to a real village, and still manages to react believably when a PC shows up may be a tall order, but I absolutely think it's a goal worth shooting for. For one thing, since many of the A-life interactions will happen in mutual isolation, the processing could be easily broken up into separate threads. Also, it's worth mentioning that this is not an all-or-nothing affair. If the equilibrium state produced at the end of several A-life generations is not exactly what you wanted, it's OK to slightly tweak the end result. The effect will still be much more convincing than the "village/dungeon/colony/factory eternally frozen in a moment until a PC triggers it."

One last bit: If you want to make AI characters seem realistic, maybe a good place to start is with Maslow's hierarchy of needs. Start there and build on that foundation. Everybody puts self-preservation ahead of other priorities, and so should AI. Ditto for all the other stuff on the hierarchy. So for example, if a fire I started destroys your hut, you will interrupt your routine to seek shelter, because this is your highest priority. For this, you may need to interact with other A-life, who might offer you a place to stay, financial help, or help with constructing a new hut - all for their own reasons, that depend on how much they like you, how likely they think you are to reciprocate, etc. If you saw me kill the miller, you will tell the other villagers, who may decide to ambush and capture you. How does a game produce that sort of behavior? Well, for one thing, planning requires some sort of awareness of expected consequences. But this should not be hard for a computer to do. It would guess that the PC would resist any attempt at capture, and it can (in the background) play out several "what if" scenarios compare their outcome to the goals of the villagers. This should show that a haphazard attack is a bad idea. Now it may look like I'm asking for a crazy amount of processing power. Maybe, but remember, we'll all have many CPU's to work with in the near future, so the ones that aren't running the game can be computing these "what if" scenarios. Also

Watch the translation pitfalls

[Jonah Hex]

Let's travel back to the heady first days of Ultima Online, when they had a spawn system that approximated real world birth-life-reproduction-death cycles. Unfortunately the player population couldn't FIND any mobs to fight, the spawn was too realistic and couldn't cope with the artificial nature of players screwing up their routines. Old spawn system trashed, new spawn system based on timers goes into place, problem solved.

The way things work in the real world is not nessesarily a good basis for a game.
Jonah HEX