CA Promises CAMPAIGN AI Improvements! Patch 8.1 - Total War: Warhammer 3 Immortal Empires

Added: 2026-05-09

[00:00:00]Welcome up boys and girls. Christine here with a discussion about the campaign AI in Total War Warhammer 3 immortal empires. Cray of Assembly has talked about that they want to make some changes to the way the campaign AI prioritizes its targets. Now, that's a fancy way of saying that they want to overhaul or improve the decision-making pro process that AI uses in a campaign.

[00:00:28]And that decision-making process does certainly have a lot lacking at the moment. Take for instance an Elbeth situation. And that's the reason I'm using her for this video. Usually in her campaign, she'll take Wizland under her control. She'll take Soland and then she will do nothing for quite a few turns until she has a border with Vlad. And usually that will happen when Vlad takes uh Sterland. The reason he's going to take Sterlin more often than not is because he's incentivized to do so uh due to the changeling weakening that specific faction. But this is one of the problems, one of the starting point issues that we could talk [music] about as an example. The idea that a faction is just going to sit there doing nothing because it doesn't want to traverse neutral territory. There's a gap between how the AI handles decision- making in terms of diplomacy when we think about uh strength rank which is one of the main determining factor uh aversion between different factions versus the strategic reality that does exist in a campaign and I think it is one of those important factors for creative assembly to actually get right.

[00:01:50]We'll see if they do so. Hopefully they would actually because it would be a genuine benefit for the game as a whole if they were to improve things from that perspective.

[00:02:05]That is one of the aspects. Another aspect is with regards to how the AI can either be ridiculously aggressive and try and take out every single salamament that they seem they think is a weak target. This is something that usually happens when you're fighting Vlad that he'll ignore Zufar because dwarves being dwarves are fairly strong especially on higher difficulties concerned that I'm talking about legendary difficulty here.

[00:02:34]And so dwarves being dwarves uh the AI is just Vlad and the dwarves are going to ignore one another. The dwarves because they're afraid of Vlad for good reason. And then Flad is going to ignore them more often than not on this kind of difficulty because he sees Sterland and Averheim really as easier targets. But he's leaving his backside undefended. And it's not just a question of Zufar. It's also a question of Unrim. If Unre were to win against AAG, that could also affect Vlad in a significant way. This is why usually when I play longer campaigns in Immortal Empires, I almost always see Vlad being wiped out. Sometimes it doesn't happen, but it's fairly rare because you need to have a situation where Unrimar all fail in order to contain the threat.

[00:03:36]And this is regardless of faction randomization. eventually, even if faction randomization, which I think is not necessarily the best way they could have handled that specific idea of like, oh, some factions are going to be better than others. This is what I actually hope that they change with Lords of End Times. If we're going to get faction randomization, it should hopefully align with the kind of endgame crisis we have enabled or endgame events that we have enabled. And maybe they just remove uh the faction randomization or all those factions all those factions of a particular race get buffed. But especially would be great for instance when we think of the confederation issue cuz this is another aspect where the game kind of fails in terms of AI campaign AI decision-m because AI factions are very rarely going to confederate one another. It happens between major factions and minor factions, especially if minor factions do end up being weak. But I've gone many campaigns with the exact same thing where no one confederates anyone in Wolf One, no one confederates anyone in Nagrant. Now I understand that the order tide and even chaos tide the constant confederations that would occur in Warhammer 2 did get annoying because they then ally each each other like Tyrion for Grim and Carl and also Mas the Mundi and Hexatal they would all ally each other but the way things are being handled right now where confederations happen very rarely and take a significant amount of time like you can go an entire campaign of 100 plus turns and see very few major factions confederate one another and that is a flaw in how the game is designed from that perspective. Maybe some force confederations. They will use force confederation mechanics sometimes, but you know, high elves are in that issu are have that issue because it's also with the seats. And then we can talk about unique campaign mechanics or racial mechanics for specific races and how those affect campaigns as well. Uh then there's also the dynamic of like a faction starting a war and going in a direction and then just stopping there.

[00:05:55]For instance, Alfarian usually will take southern rest or always will take southern of rest, but more often than not, he'll just stop there with multiple stacks doing nothing while his north is threatened by Nikkari. He can actually lose to rest because of that because his armies are just sitting there doing nothing. There's a lack of concentration of force. There's a lack of proper utilization of resources, especially again on legendary. Uh then we could talk about how due to the various cheats AI has in terms of construction costs, upkeep cost, recruitment costs I believe as well that you end up in a situation where uh certain factions end up with huge piles of cash and they're willing to give that to you or almost all of it to you for just a declaration of war you don't really care too much about.

[00:06:48]And 7,000 is not even the most ridiculous number. Like we can think of uh green skins where if you region trade with green skins, if you regen trade with say Grimgore, I'm not sure how well he's done over here. Uh not necessarily too well, but sometimes you do get like Reese's having tens and tens of thousands of gold, Nurgal factions having tens of thousands of gold. um and other factions as well having a huge pile of resources uh that they're not necessarily utilizing properly. They recruit a certain number of armies then they stop. I'm not quite sure what the decision making process is regards to that but I think the AI should uh utilize things better in terms of resources. Uh the wood elves are one exception in that they tend to have very little in terms of actual money sometimes. Like for instance here, dry only has 331.

[00:07:54]But that's because the wood elf economy is based on buildings with growth. And so the AI is generally going to use that growth for some other buildings. For instance, I believe this one is also using growth. uh and instead of building up economic buildings, but that's more of a wood elf situation. But without that money, they actually end up being more limited. And also, the armies that wood elves do usually recruit tend to be lackluster compared to their potential because they're not prioritizing the things that they should be prioritizing in terms of building uh building up properly in their empire. That's all I had to say. Christine signing out. Don't forget to subscribe, like, and enable notifications. It's obviously a technical subject, very technical subject, but it does touch on the core of the game. I'm not sure exactly how far Crave Assembly is going to be willing to go with regards to it. But, you know, so far we've seen them make some changes with regards to AI prioritization and then just kind of leave them there for a while. Hopefully 8.1 will be a major change with regards to campaign AI prioritization and not just a minor one.