I believe that we will see more games that have AI-ish systems. Randomization in games like Don't Starve is one of the mechanics that keeps the game going.
A Kalvin-ball AI would be a pretty interesting novelty but I think good games are usually focused on simple rulesets. Not sure an AI is really needed there. Maybe you just mean hyper tuning drop rates and and modifiers and such?
I'm sure AI will eventually have a huge impact on the art, narrative, and engineering of games, though, so maybe you're correct that will bleed into game design as well.
Game designers usually have to do tons of playtesting to find all the broken strategies in their games. As self-learning AI becomes more accessible, it'll give designers much quicker feedback. I'm curious to see how this will affect the games we play. Maybe someone will discover a new game as deep as Go and Chess!
That's hardly AI. It's just algorithms. Parametrized, using a probability distribution to randomize the environment, user programmed algorithms.
EDIT: Well, if this AI in game development, then so be it. But it's still not AI in the generic way, IMHO. To downvoters: I didn't contest the fact that they don't work. I do enjoy this games a lot and I praise the programmers that make such games work.
I wonder if the AI overlords will come up with strategies that look completely non-sensical to humans but are otherwise undefeated. I'm think of the Alpha Go Zero style reinforcement learning.
I'm struggling to see how to run thousands of games programmatically in their documentation though
My optimism is mostly for increasing the number of creators of games --> best games get even better, at the cost of introducing more games, some of which may not be good. Marketplaces being the winners is a symptom of the players being the ultimate beneficiary of increased competition (and thus a surplus of both good and bad games).
I actually don't see this as an AI to replace, but AI to enable more people to create games. So ultimately, it still expresses the desire of the creator, who is human.
This will be good for games moving forward due to the meta changing for players as the AI adapts to their tactics and vice versa. Lessons learned from this can then be applied to other areas. And as an added bonus it creates more interest in AI research.
I agree, the AI angle has promise. I think what I'm getting at is that once I get a sense of "infinite" in a game it starts to feel very pointless to me.
I suppose I'll see if that changes in a few years once AI-driven game content becomes more prevalent.
Yeah, in the end it's all if-this-happens-do-that under the hood. And that's important because game AI must be deterministic. Otherwise reproducing bugs would be impossible.
I would say any game that includes a lot of uncertainty, subjectivity and random chance would mean that an AI would not usually have an edge over a human. Maybe games like diplomacy, magic the gathering, pandemic, Gloomhaven, maybe poker (?), backgammon (?) etc.
On a sidenote, I think it’s interesting that we’re a point where it’s starting to get hard to come up with games that humans can beat AI at.
Designing a game around PvE is hard though. It's difficult to create an AI challenge that's interesting and can simultaneously challenge extremely good and mediocre players without feeling impossible.
I think the future of PvE games is going to be in AI. If game developers could put a DeepMind-playing-StarCraft level of AI into games without immense cost, then it's going to change games quite a lot. It wouldn't be perfect, but you could have AI offer a challenge to players for longer without feeling unfair. (It's still AI that doesn't learn based on the player though, which means that the players would eventually figure out the puzzle of the AI.)
that's a good argument. part of the fun is learning the AI behaviors. and in modern games it honestly would take a while for that to get boring, the AIs certainly are getting more complex
One of the most interesting takeaways from the post game interview for me was that the AI can be very stupid if you just blindly throw it in a self-play setting but with clever use of randomization (modifying power levels) and action restrictions (for example, only allowing the agent to spend an anti-invis item when a nearby enemy goes out of sight) it is possible to provide better learning opportunities for the AI.
It will be interesting to see how it works. The biggest issue I can foresee is that a program doesn't have a limit on how much it can manage, given the computation time. If one AI is even a little better than the next best, it could quickly dominate the entire field without some limitations on growth. Something like warcraft 3's upkeep system would be good for ensuring that the game doesn't become boring even if one player is clearly better than the rest.
AI doesn't have all that much application to gamedev - people think "videogames have lots of AI", and don't realize that "game AI" and "stable diffusion/GPT AI" are about as related as Java and Javascript.
Game "AI" is meant to be fun to play (and win) against, they're not meant to be smart; that's why zombie games are so successful. Most "game AI" are finite state machines, throwing a neural network at the issue would be absurd overkill.
I'm sure there will be some AI applications in games (like procedural world generation or such, perhaps) but it's not the obvious connection that most people think.
When I was in grad school, I was working on general game playing AI. Unfortunately, I was in a "pure logic" research group, founded on the old-school AI principles that believed AI could be derived from deterministic logic.
Of course, this limited the games that we could simulate to purely deterministic games (checkers, chess, go, etc.). Any games that included an aspect of chance required a hack like a "dice player" or a "deck player" that would add the random aspects of the game. Of course, this led to other problems, since the engines would try to calculate the current state of the game based on the "optimal" play of the random player.
This is a much more interesting approach, and I imagine will prove to be far more useful.
reply