D&D (2024) Command is the Perfect Encapsulation of Everything I Don't Like About 5.5e

[Warpiglet-7]

Codification I want is more like examples and benchmarks that help the GM to extrapolate consistently, not exhaustive and limited lists of all possible things.

This 100% what I want. Clear examples of how/when rhe dm can use the skill.

For example a DM might call for a stealth roll if…

Then follow hints for how to set the DC.

I would not even mind a few options—-meaning the group’s DM might choose to use it in A, B or C way with some codification in there.

I know people like certainty and things to be systematic. I do too frequently but I also like the rpg wor to be open too.

It’s a balancing act. There are editions that went to far in either direction for my personal taste but at least I now know more about what appeals to me!

 

[mellored]

AI shouldn't have a problem running a single word command.

Results might not be what you expect, and occasionally just nonsense, but chat bots will always be able to provide some kind of answer.

Here's a quick AI answer


ChatGPT
In a Dungeons & Dragons game, if a cleric successfully casts the Command spell and commands the king to "resign," here's how it might play out:

Immediate Reaction: If the king fails his saving throw, he might react with confusion or fear, feeling compelled to step down from his position. This could lead to immediate chaos in the kingdom as he announces his resignation.

Political Fallout: The consequences would likely be significant. The kingdom's court, nobility, and subjects might be thrown into turmoil. The party might face backlash or have to deal with the political aftermath of the king's resignation.

Role-Playing Opportunities: The players can now navigate the fallout from this dramatic event. They might have to deal with power struggles, address the needs of the kingdom, or deal with any immediate threats that arise from the sudden change in leadership.

 

[Ruin Explorer]

I think people making assumptions about changing things so an AI can run a game is a bit farfetched. There are quite a few things in D&D that just don't mix well with automation and likely never will. If there were an AI that could run a game they could just limit what the AI does and nobody would be the wiser. Yes, the AI only tells you to do the things that are on a short list but most players will only use a limited option as well.

Occam's Razor tells me that it's just something they find people don't use very often because it's so open to DM interpretation and it's too vague for a lot of people. That, and it's only a first level spell that in the old version needed to be understood and interpreted by the target. The new version is not the target interpreting the spell, it's the caster invoking something from a list of options. It's actually more powerful for the majority of players because the target doesn't need to understand the language.

Yup.

I don't expect D&D to go full AI DM, much as WotC leadership (not Crawford/Perkins level, above them) 150% wish they could do that. D&D just is too DM-centric and too much about the DM thinking up what happens next or interpreting stuff. That's actually a good protection from automation (I say this as someone who works in automation, note!). If you actually need to stop and think about stuff, AI is not the answer, generally-speaking. AI is okay for generating stuff (not great), and great for summarising stuff, but coming up with solutions to situations/problems? Unless a lot of people already wrote solutions/resolutions and put them on the internet, nah lol

I think there are some solo-oriented RPGs (which you can also play in a group, some very well), which are already trying to have a sort of reflexive or automated DM where AI could be more of a replacement, but it would it need specialist training, you couldn't just drop ChatGPT or the like in there, and I think for the short term, that kind of specialist training would be outside the budgets of the sort of RPGs where it might work (barring a spectacular Kickstarter focused on that).

In short, we may well see a TT RPG designed specifically so that having an AI DM works, but that RPG will not be D&D - and probably not any RPG that exists now,

 

[Ruin Explorer]

The demo they had included automatic saves for a breath weapon and damage. But just because some things can be automated, there is no goal of automating everything.

Yeah I'm not surprised - that's a little more than I'd been told but it's still in the same general zone as other VTTs, it's not a serious automation attempt.

 

[Warpiglet-7]

Yeah part of this is hard for me because I don’t have the book in hand so I can’t see the phrasing but I gotta be honest, I’ve never understood some of the contortions some folks have proposed with existing 5e rules. It could be a nothing burger for all I know.

I am 100% withholding judgment until I can look at a book.

I keep going back and forth about it…

So far I am likely to get DMG. Will know more about PHB and MM later. I don’t know how telling this spell really is…just one little data point.

 

[Oofta]

Yup.

I don't expect D&D to go full AI DM, much as WotC leadership (not Crawford/Perkins level, above them) 150% wish they could do that. D&D just is too DM-centric and too much about the DM thinking up what happens next or interpreting stuff. That's actually a good protection from automation (I say this as someone who works in automation, note!). If you actually need to stop and think about stuff, AI is not the answer, generally-speaking. AI is okay for generating stuff (not great), and great for summarising stuff, but coming up with solutions to situations/problems? Unless a lot of people already wrote solutions/resolutions and put them on the internet, nah lol

I think there are some solo-oriented RPGs (which you can also play in a group, some very well), which are already trying to have a sort of reflexive or automated DM where AI could be more of a replacement, but it would it need specialist training, you couldn't just drop ChatGPT or the like in there, and I think for the short term, that kind of specialist training would be outside the budgets of the sort of RPGs where it might work (barring a spectacular Kickstarter focused on that).

In short, we may well see a TT RPG designed specifically so that having an AI DM works, but that RPG will not be D&D - and probably not any RPG that exists now,

I wouldn't be surprised by AI assistants, but the AIs we will likely have for the foreseeable don't really "think". An AI that helps with planning an adventure, gives suggestions on possible encounters? Sure. Run a game? That would require something along the lines of the singularity.

 

[Kobold Avenger]

Someone pointed out that Command if it allowed any single word, might be even more powerful in some of the non-English international versions of D&D.

 

[James Gasik]

Personally, I'm on the fence here. Removing the need for the target to understand you, while admittedly a PITA, is kind of nonsensical. I'd add that it weakens the point of learning different languages, but that ship has sailed into the sunset, lol.

OTOH, I'm so tired of people coming up with ridiculous commands that are too OP, I have no idea how to interpret, or are just in poor taste.

Like telling an armored character to strip or m*st*rb*t*, ffs. I've had nothing but problems with open-ended spells in the decades I've been gaming. Illusions, suggestion, wish- these are always moments where player expectations and DM expectations collide.

The player is always trying to see what they can get away with, and the DM is constantly trying to make sure the game's balance doesn't implode- the end result is a lot of "meh".

Honestly I wish suggestion was simply removed from the game- the constant arguments about what is "reasonable" made it not worth casting (but somehow it always works out for the NPC's, lol). But removing that word from the spell has just made it worse, IMO.

People on both sides of the DM screen will end up attempting to abuse it, and it's going to be fun for nobody.

 

[Minigiant]

Again I think D&D the biggest issue is that D&D has a broad audience.

Heavy power free-form abilities with major encounter change is just not feasible.

For a niche game with a smaller targetted audience, sure.

 

[Shardstone]

Someone pointed out that Command if it allowed any single word, might be even more powerful in some of the non-English international versions of D&D.

That's totally ok IMO.

Spells should be fun, not little mechanical packages. I'm ok with mechanical packages, but being able to interpret spells to change the output of them is something that I think is core to roleplaying "magic" in any kind of context. It requires a human to moderate this output so it doesn't ruin the experience of the game, but that's why we play TTRPG over boardgames. There is a GM who I trust to make a good decision, and who I'm willing to go with their arbitration. That inherent trust is core to the experience as well, and I feel like designing around that trust makes for a more stifled game.