Is it true that the AI cheats?

[Sapare]

If Glavius is to be believed the AI spawns itself a good amount of Energy credits to then turn them into other resources through the market.

I had not heard of this before and considering the reaction last time the AI had undisclosed bonuses, it would strike me as brazen to attempt this again.

So, here are my questions.

Does the AI actually spawn itself energy credits?(effectively any resource its willing to buy on the market)

Is this intentional or a bug?

Is this commonly known or underhanded?

Does anyone care, considering the state of the AI even with those tools?

Maybe someone with modding knowledge can dig through the code, or people can run observation games and report their findings. If you go back in my posted topics you see I had some suspicions early on from 2.2 but in the end I am only going by what a well respected moder said on his own mod page.

Edit: And my concern is primarily for the non cheat difficulty levels of AI. In my tests it seemed like the AI spawns these resources even on Ensign, but I admit Glavius never made clear what difficulty he was referring to.

 

[SirL]

Yes it cheats a lot and you can even see it in description in difficulty. Even on commodore the amount of cheating is unbelievably high, but it is so bad that I always win wars with 3-4 times worse fleets. Even with all those cheats it will eventually become pathetic to you, as it can't keep up economically

 

[GC13]

You can see a big Steam discussion about it here. I haven't bothered testing it, but if Glavius (who is basically the AI guy) says it I have no reason to doubt him. Stellaris's AI is incredibly feeble, and it wouldn't surprise me at all that a crutch like this was either a bug and missed in testing because even with it the AI can't effectively defend itself, or was intentionally added in the hopes that it would maintain the illusion of normalcy until some day when they could get things back under control with the AI.

 

[Masoz]

Hah, so much for the "economic warfare" they marketed. The market only applies to the player.

There's no way this wasn't a crutch, imo. The AI is not developed enough to build their own functioning economy, so they gave them unlimited alloys so that you at least have something to shoot at.

Don't forget that pre-release testers warned that the AI was incapable of building fleets larger than 2k by endgame. I have a sneaking suspicion this "unlimited resources" behavior was programmed in last-minute to avoid that.

 

[sti]

I'm not sure why people are always so shocked at this kind of thing that AI in every 4X game is "bad". These games are simply too complicated to keep up with any reasonable player.

 

[Little Red]

Yes, the AI has and always will cheat. It's essentially a given in plenty of 4X games. Even on "normal" difficulties, the AI will cheat in some manner to augment the fact that it's no easy feat to have an adaptive AI for such complex systems.

I personally feel that Paradox should do what plenty of developers do and put modded fixes into the base game. With credit, of course. Heck, they might even hire Glavius to work at Paradox if they do so!

 

[GC13]

Heck, they might even hire Glavius to work at Paradox if they do so!

That would reduce profits.

 

[JOD]

I don't mind the AI cheating with resource gain, I play on Admiral to Grand Admiral. I accept that it's cheating and don't really care. I'd go so far as to say I actually prefer it that way.

I *do* however mind when the AI is clearly playing by a different rule set or ignoring mechanics that the player must engage with.

I especially mind when the AI does the aforementioned thing and the developers do not disclose it to the the player.

The way the development diaries were said, it was implied that the market worked the same for the player and the AI. AND that a crafty player could hypothetically exploit weaknesses in the AI economies by driving up or deflating prices with resource dumps and mass buying.

If the AI really does buy things off the market for free, then as a player I feel a tad bit manipulated by the developers.

 

[Jman5]

I'm not sure why people are always so shocked at this kind of thing that AI in every 4X game is "bad". These games are simply too complicated to keep up with any reasonable player.

It's really sad that so many people believe that 4X AI are incapable of being anything but bad at the game. So many of these 4X games have AI mods that prove this wrong again and again. It's not that they can't make the AI better. They just don't think it's worth their time because most people are perfectly content fighting a massively broken AI. At worst they believe it's counter productive because they want the player to win.

If suddenly AI competence became a major selling point for the gaming industry, you would see massive improvements in vanilla AI. They're smart people over at Paradox, it's just not a priority to make an AI that will beat you unless you play pretty well.

 

[Ariphaos]

I don't think the AI getting free resources from the market is remotely intentional. It takes a fair amount of game knowledge to handle multiple hostile AIs with 40-50 ship fleets ~20 years in.

I'd be fine with a difficulty-scaled market discount for the AI though. Maxing at say, 50% at Grand Admiral.

 

[Foefaller]

It's really sad that so many people believe that 4X AI are incapable of being anything but bad at the game. So many of these 4X games have AI mods that prove this wrong again and again. It's not that they can't make the AI better. They just don't think it's worth their time because most people are perfectly content fighting a massively broken AI. At worst they believe it's counter productive because they want the player to win.

If suddenly AI competence became a major selling point for the gaming industry, you would see massive improvements in vanilla AI. They're smart people over at Paradox, it's just not a priority to make an AI that will beat you unless you play pretty well.

Do those AI mods really make them play "better," or just make decisions more optimally?

I know those sound like the same thing, but there are tons of little choices that can be made to get a little more if a resource, or get things done a little bit quicker, that make the AI more of a challenge, but don't actually make the AI "smarter," or more specifically, more situationally aware of what a good decision is.

One example I've seen an actual AI dev talk about in relation to 4X is tech trading in Civ games, back in Civ 4. It's easy to tell the AI that techs have a value, that you should try to get something close to that value in return, and that you shouldn't trade with someone who might use it to beat you, but you can't really tell it (at least with the understanding of AI at the time) how to know when it is a good idea dump technologies on a civ that has 0 chance at catching up to them for far less gold than it is worth in order to use that extra gold to rush an important wonder or building, but that is a value judgment a human player can and does do all the time, to great effect.

I also remember back when someone made an AI that could beat a grandmaster at Go comparing the number of moves that computer had to consider compared to what the AI does with Stellaris. While it was some pretty rough napkin math, I came up with pre-2.0 Stellaris having more than twice as many possible combinations of purchases you could make with just your starting minerals and energy than are possible in an average turn of Go, and that is arguably when the game is at it's simplest. That inevitably ends up ballooning that to much, much, higher, with a smaller proportion of "duplicate" actions that are basically the same thing in a slightly different order, and the Stellaris AI usually doesn't have a super computer's processing power to work with.

That said, I have no doubt the AI will at times deliberately pick the less efficient option for reasons that have nothing to do with roleplaying a type of empire or limits to the software, but I'm also more likely to believe that, if this cheat is intentional, the most likely reason is as a band aid fix until they get the AI to work at the level of competency they want without imploding.

 

[wingren013]

Yes this is true, or at least it looks like it is. The AI's energy does weird things if you watch them in observer mode.

It's really sad that so many people believe that 4X AI are incapable of being anything but bad at the game. So many of these 4X games have AI mods that prove this wrong again and again. It's not that they can't make the AI better. They just don't think it's worth their time because most people are perfectly content fighting a massively broken AI. At worst they believe it's counter productive because they want the player to win.

If suddenly AI competence became a major selling point for the gaming industry, you would see massive improvements in vanilla AI. They're smart people over at Paradox, it's just not a priority to make an AI that will beat you unless you play pretty well.

The type of problem presented by 4X game is a particularly difficult kind for AI. Like the poster above said, it's possible to identify specific behaviors that the can be changed or added to make the AI play more optimally, but actually getting it to have more to its actions than the immediate is a massive problem.

 

[Sapare]

My central concern is this, if the Ensign AI cheats, without disclosure. Then Paradox, the same team at paradox, lied to us AGAIN. About the same topic that broke a lot of peoples faith the first time around.

There is making a mistake once because of misunderstanding the situation, and then there is doubling down a 2nd time while preaching how you learned your lesson.

If the AI is so blatantly cheating on Ensign(and below) then I am seriously upset, and no, "it was a bug, totally accidental" doesn't cut it here, not when you already got caught with your hand in a cookie jar not a year ago, with a cheat that exited in the code for YEARS. They made a promise this wouldn't happen again, whether the AI needs to cheat or not is irrelevant, you can't put in cheats and not disclose them after promising otherwise.

IF this is true, I don't see how I can support a company that repeatedly lies to its costumers. I adore Paradox games but selling your game under false pretenses is rotten business practice.


Edit: I do forgive it as a bug if its in the beta only, has there been any definitive answer if this is a beta branch issue or a general 2.2 issues? The beta branch is not an "official" release and while a fishy bug, i accept it. If its 2.2 I don't accept a bug excuse, someone on prohibition isn't accidentally drunk driving.

 

[Scorpio_Shirica]

I wouldn't mind except that in the test version, once the global market appears, it completely breaks the alloy market. The AI continually spam buys alloy so if you wanted to do a trade heavy civilization that bought all its alloys, you may as well forget it.

 

[Glavius]

AI seems to be able to cheat 650 energy credits when reaching 0 credits with -49 energy, +6 minerals and +0 consumer goods (machine empire) monthly and a mineral stockpile of 49
AI that reaches negative income/ empty stockpile on minerals, energy, food and consumer goods cheats 1.1k energy credits
Link - Steintim

Every single copy of 00 defines has the following line in it:
AI_POP_MODIFICATION_DISCOUNT = 0.25 # AI pays this much (of base) for gene/robomodding projects

The AI seems to frequently have energy appear from thin air. I do not believe Paradox is using an event to create these magical resources. These are hard coded cheats as far as I can tell.

The AI does seem to sell on the market, but the numbers showing resources sold stay stuck at zero. I think this is a bug and not a cheat.

I do believe the AI buys from the market like a normal player. However it frequently pays using cheated currency.

It is my goal to get the AI economy working in such a way that it will never have to cheat resources into its stockpile. My mod is currently an optimization mod and not yet making the AI actually better at long term strategic planning. The modding tools however make it possible to go far beyond simple optimization of behavior. However, 2.2 hasn't been out long enough for me to get to creating actual AI personalities tied to the various types of governments/traits/traditions.

 

[TwoToTango]

The purpose of the AI is to give the impression of a competing space empire and to give the player a challenge that is fun, that is not too too easy and not too hard. Why does it matter how this is accomplished? The only thing that matters is how the player perceives it when playing the game. You want the experience of the other empires playing by the same rules but it doesn't matter if they actually do behind the scenes. It wouldn't matter to me if the AI was basically braindead and just spawning buildings, ships and fleets out of nowhere in a way that is designed to give the impression of producing them. It certainly doesn't bother me that it is given extra resources to compete with a player. I don't see how it would bother anyone.

Until we have actual AI then AI in a game like this will never be able to compete with a player that knows what their doing and is playing competetively. The solution is to make it play by different rules to some degree. That or you could take away the pause button, giving an edge to the computer that can think much faster than us. You can't both be upset that the AI cheats and that it is not challenging enough, that is completely counterproductive.

 

[Pooks1]

Well this topic does explain something I observed in a few of my games.

I haven't been able to play 2.2 a lot (yaay for holidays to come!) but it was really weird to me that the AI was able to go from equal to overwhelming in fleetpower within just a decade or so, all the while me being at my max fleet capacity. Bear in mind that this was all early game, within about 30-40 ingame years passing and on commodore difficulty. After that I tested it on captain difficulty and same thing, which even more shocked me. The AI never had that much of a boost in earlier versions, and needless to say they attacked me at once being despotic slavers and me being a pacifist megacorp.

 

[Ariphaos]

They clearly don't need AI_POP_MODIFICATION_DISCOUNT any longer. The AI is spamming tiny little custom races all over the place.

The purpose of the AI is to give the impression of a competing space empire and to give the player a challenge that is fun, that is not too too easy and not too hard. Why does it matter how this is accomplished? The only thing that matters is how the player perceives it when playing the game. You want the experience of the other empires playing by the same rules but it doesn't matter if they actually do behind the scenes. It wouldn't matter to me if the AI was basically braindead and just spawning buildings, ships and fleets out of nowhere in a way that is designed to give the impression of producing them. It certainly doesn't bother me that it is given extra resources to compete with a player. I don't see how it would bother anyone.

Until we have actual AI then AI in a game like this will never be able to compete with a player that knows what their doing and is playing competetively. The solution is to make it play by different rules to some degree. That or you could take away the pause button, giving an edge to the computer that can think much faster than us. You can't both be upset that the AI cheats and that it is not challenging enough, that is completely counterproductive.

Right now, facing the AI with this infinite resource generation ability can be insanely difficult in some scenarios. In one of my devouring swarm games, I was stuck between two AIs, who promptly allied. Nothing I could do, short of cheating myself, could win the war against both of them - they were fielding nearly a hundred ships to my ~twenty.

 

[FlyingPhoenix]

The purpose of the AI is to give the impression of a competing space empire and to give the player a challenge that is fun, that is not too too easy and not too hard. Why does it matter how this is accomplished? The only thing that matters is how the player perceives it when playing the game. You want the experience of the other empires playing by the same rules but it doesn't matter if they actually do behind the scenes. It wouldn't matter to me if the AI was basically braindead and just spawning buildings, ships and fleets out of nowhere in a way that is designed to give the impression of producing them. It certainly doesn't bother me that it is given extra resources to compete with a player. I don't see how it would bother anyone.

The issue is that this is a strategy game, that a valid military strategy is to snipe the opponent's economic and production base, and so if the AI isn't playing by the same rules as the player, that strategy becomes invalid. Which is the point another poster made in a post now deleted - the AI should be fun to play against. Having the AI blatantly cheat and not follow the same rules as the player often doesn't post an interesting fun challenge - because the AI's economy is based on console hacks, it means that the interesting and fun strategies the player can employ are null and void.

 

[Aleriez]

They clearly don't need AI_POP_MODIFICATION_DISCOUNT any longer. The AI is spamming tiny little custom races all over the place.

Actually the discount is necessary because the AI doesnt know that it has to pay the base research value too and it doesn't understand that this stops research...
I assume that the AI does pop modding as soon as it is available, no matter the cost. E.g. modding one race for 32 years, with the discount its only 8 years...

IMO they should get a cooldown on pop modding that prevents them from doing it too often (if thats not already a thing...)