I’m not sure I really buy the argument that this could be better for several reasons.
-
The implicit guardrails these companies are going to add which will complicate things.
-
Numerous game-breaking states because you’re risking a more traditional Dungeons & Dragons Dungeon Master problem where your party somehow has failed to ask an NPC the right kind of questions or even consider that they might have information relevant to the campaign. How do you get this information across if the player isn’t somehow prompted to attempt it?
-
Baldur’s Gate 3 only came out about 4 months before this article. It exists as a counterpoint to the idea that pre-scripted dialogue can reduce replayability. Only if you approach it like Bethesda and you refuse to block your players off from content. I don’t think Todd Howard realizes how that philosophy hurts replayability because you can always complete every questline every playthrough. While in a game like BG3, large amounts of the game are locked off based on your character, class, party and choices you make. Certain things you have to replay the whole game to access. There’s mountains of replayability in their world of pre-scripted dialogue. It’s rather no other company has been willing to put the kind of time, money, effort, and production quality that something like BG3 demands.
I think this AI stuff is a cheap cop-out that uses way too much energy for a weak result. Instead of making better games the system spec requirements will either become insane or all games will be delivered via streaming platforms. They’ve maxed out graphical fidelity but still need excuses to use “better” hardware. Better game design achieves the same result without the vendor lock-in and absurd hardware/power demands.
The implicit guardrails these companies are going to add which will complicate things.
That’ll just have to be part of evaluating whether a game is “good” or not, I guess. If game companies hobble their NPCs with all sorts of limitations on what they can talk about then it’ll harm the reception of the game and drop its metacritic score.
I do see some interesting hurdles that were likely never imagined when the rules were written. How do you come up with an ESRB rating for a game where you don’t know what topics your NPCs might talk about or what sorts of quest lines might ultimately be generated?
Numerous game-breaking states because you’re risking a more traditional Dungeons & Dragons Dungeon Master problem where your party somehow has failed to ask an NPC the right kind of questions or even consider that they might have information relevant to the campaign. How do you get this information across if the player isn’t somehow prompted to attempt it?
That seems like something that an AI-driven game might actually be better at, if properly done. The AI could review the dialogue the character has participated in so far and ask itself “has the player found out the location of the cave with Necklace of Frinn yet?” And if it sees that the player just keeps on missing that vital clue somehow it could start coming up with new ways to slip that information into future dialogues. Drop hints and clues, maybe even invent a letter to have delivered to the player, that sort of thing.
Whereas in a pre-scripted game if a player misses a vital clue they might end up frustrated and stuck, not knowing they need to backtrack to find what they overlooked.
I think this AI stuff is a cheap cop-out that uses way too much energy for a weak result.
If the games using AI aren’t good then they won’t sell well. This is a self-correcting problem.
The guard rails already killed an experiment of it, Square Enix did a remake of an investigation game with LLM but the model was completely useless, it used to be the lowest rated game on steam.
-
I’ve used an LLM chat bot to play D&D with and it worked pretty well. At least for running the game and making shit up based on the campaign I wrote it to run. It wasn’t really much different than playing with real people; not least of all because it wasn’t perfect. It would make mistakes, just like a real person. It would misinterpret the rules, like a real person.
So I think this kind of “AI” could be used pretty well for games. However, from how these major companies have been using it, it won’t. Because they’re gonna try to get the entire fucking story written by it, and that’s when it’ll fuck up.
It’s decent at making up new shit on the fly creating dynamics never really found in video games, but it isn’t always necessarily congruent or logical. A game like Dwarf Fortress could benefit from it, but a game like The Last of Us would suffer from it.
I tried that as well, but for me it was like being 10 again: -you meet the bandits -I, the lvl1 player kill them all -OK -I just remembered my party had a necromancer, raise the corpses -sure thing -I march around with my undead army and murder everyone who is in my way -This game is about creativity and cooperation -Not today -OK
9 months old.
I hadn’t read it before, and I thought it was interesting, and the article is still as relevant as it was back then. I thought many others missed it too. It’s also pretty well written.
Need me a cyberpunk detective game with real verbal interviews.
What’s a tortoise?
