While he is acknowledging that AI needs to be used responsibly the things he is talking about using it for are very troubling IMO. The statement is clearly not a 'we will use AI grammar checking for human generated story lines' but rather that AI will be in some way involved much more fundamentally in game development which IMO is a reversal of their prior stance and very quickly gets into exactly the bad stuff nobody wants. I don't think we will necessarily see AI images or entire books, they won't likely do that for copyright reasons at least right now but I would like to see much more of a firewall than what he is describing. If I want to consume material made by actual artists/writers not what an AI scraped off of actual humans and regurgitated. Sure some AI concepts might be fine but that is not something I am willing to pay for.
The issue with the rules role playing wise is that they do a very poor job of arbitrating non-combat actions and anything that is not epic fantasy style play. The binary on/off skill proficiencies for example do not allow for a character to reflect that they are a bit better than average cook but not a well trained chef. But the biggest problem is the inherent level of randomness when you have a 20 sided die and a +2 - +5 proficiency bonus and a -2 - +4 or so stat. This just does not reflect the actual difference between someone who is highly talented or highly skilled at something and someone who is not. That is made worse by going all in on advantage/disadvantage rather than using more numerical modifiers. The result is a lot of work on the DMs part to make up for silliness the system introduces. Take an experienced ranger and a wizard with no survival skills. Give them both a moderate survival skill situation and the ranger might have +3 stat +3 bonus in survival and the wizard might have +1 and +0. The difference is a 25% better chance that the ranger succeeds when it should be a much higher chance. So we patch things up with passive scores where you assume a 10 was rolled and whatnot but you still end up with the wizard pulling off some insane feat of survival without having any training and the trained person failing.... IF you use the system to support your role play. So it falls exclusively on the GM to arbitrate who is able to do what without system support.
Even in combat the system is all about the epic fantasy arc where an epic character (high level) can survive and dish out insane amounts of damage compared to an 'ordinary' human. That is fine for an epic fantasy system so I have no issue with it but it is a limitation. The lack of support for just about anything outside combat is a real problem though.
Reading this leads me to believe that you haven't actually looked into how D&D handles ability checks and skills as a system in general, so here's some info to expand your knowledge:
Degrees of success and failure exist within 5th edition and have done since 2014. Generally it takes the form of if you succeed or fail a check by 5 or more, there's an additional effect. There are also rules and suggestions for increments of 5. This is present in both the 2014 and 2024 DMG
Difficulty in D&D is (in my experience) actually quite easy to calculate. The 5/10/15/20/25/30 DC scale is based on something being very easy/easy/average/hard/very hard/near impossible for the average person. This is a very useful rubric as guess what, you and I are the average person. So I can go with "Would myself and the people I know find this task easy/very easy/hard/near impossible" and then set the DC from there
You seem to be equating the ratio of volatile variable (dice) to nonvolatile variable (modifier) when determining "randomness". Not only is this not a good metric of randomness, your numbers are off. A level 1 character in a class that doesn't get expertise in something will typically have a +6 in a key skill; +2 proficiency and +4 from their primary ability score. That means they have the following probabilities of success based on typical difficulty - Very Easy (DC 5) - 100% - Easy (DC 10) - 85% - Moderate (DC 15) - 55% - Hard (DC 20) - 35% - Very Hard (DC 25) - 10% - Near Impossible (DC 30) - 0% That's a very logical spread that isn't super random and that's for a level 1 character with no expertise. Let's compare this to a character making the same ability check with a non-key skill (no proficiency, +2 modifier) - Very Easy (DC 5) - 90% - Easy (DC 10) - 65% - Moderate (DC 15) - 40% - Hard (DC 20) - 25% - Very Hard (DC 25) - 0% - Near Impossible (DC 30) - 0% Yes, there's an overlap of being able to do things, but that's intentional and (IMO good) design. Having it so only one character has a chance of doing something because they have the right skill is not great design in my opinion (something I touched on earlier). But if you think that's good design, the rules cater to that too because there is mention of only allowing characters to attempt ability checks if they have proficiency in the relevant skill, which would for example solve this below example:
Give them both a moderate survival skill situation and the ranger might have +3 stat +3 bonus in survival and the wizard might have +1 and +0. The difference is a 25% better chance that the ranger succeeds when it should be a much higher chance.
There are the tools you want at your disposal in the rules. You've just got to actually look for them.
Oh, one more thing. I've googled the heck out of the matter and I can't find anything saying WotC is gonna start using AI to generate content. I've seen stuff about them using AI tools in their process for stuff like information management, database control, etc. And talk of building AI tools to assist DMs in forms such as rules engines and bots that can help you play. But zilch about generating content—everything has been very much "we're going to keep paying real humans to write the rules, make the art, and create the game"
Realityhackphotos, remember that ability scores are typically only made when there's a meaningful risk of failure or there's a time crunch. If a character with a +5 athletics bonus takes their time to carefully climb up a DC 15 wall, they will climb it successfully.
Realityhackphotos, remember that ability scores are typically only made when there's a meaningful risk of failure or there's a time crunch. If a character with a +5 athletics bonus takes their time to carefully climb up a DC 15 wall, they will climb it successfully.
Yes that is true. And that papers over the problem in many situations. Unfortunately it doesn't fix them all. When you have an opposed roll, or a contest between two characters, or meaningful consequences for failure that doesn't fix the issue. But it does reduce how much you encounter it. It also doesn't really show the difference in what wall a highly skilled climber can climb carefully vs. what a novice can IMO.
To be honest I was in a bad mood when I made the OP and upset about the Hasbro CEOs comment and ranted rather harder than I intended to.
IMO 5th ed has some welcome improvements over 3.5 but they also went the wrong direction with some stuff, I understand it was largely a result of their attempt to simplify things but IMO they went too far and in the wrong direction entirely. Skills and stats should IMO have been made more meaningful not less, and the skill list expanded not trimmed. Yes there is complexity in that but it also opens up a lot of options for actively encouraging characters to be unique and fully reflect the players vision for them. I also think advantage / disadvantage is a fine mechanic but it is currently overemphasized at the expense of numerical modifiers that are sometimes just a better way to handle things.
Yes, there's an overlap of being able to do things, but that's intentional and (IMO good) design. Having it so only one character has a chance of doing something because they have the right skill is not great design in my opinion (something I touched on earlier).
I agree that *Some* overlap is good design I disagree with the amount of overlap. The difference between an untrained climber a trained climber and a top expert climber is a lot higher than what the system simulates. There absolutely are lots of things the untrained climber has zero shot at accomplishing that the expert climber can achieve reliably. And I don't have any problem with the game reflecting that. The issue with not allowing the skill check if they don't have proficiency it still does not allow for the wide range of abilities. Sure you can say nobody who is untrained can climb the wall, but that doesn't address the difference between a barely trained untallented person and one with extensive experience. There is expertise which provides a little bit of a patch for this issue but IMO it is inadequate.
More than that I also don't like the skills all being binary proficiency/non proficient. IMO this really limits the ability of the system to support a given characters traits. You cannot in the system simulate numerically being a little trained in x and extensively trained in y the way you could in previous editions. Outside of combat where there are a ton of different rules interacting to create much more variation in ability levels the system itself struggles IMO to support using rolling dice to determine outcomes.
ETA: And using advantage/disadvantage very heavily instead of numerical modifiers also creates very binary situations. You can't get extra advantage from additional favorable circumstances.
This sounds like your objection lies in the numbers not massaging the ego. The game is perfectly set up to handle success and failure of epically heroic individuals attempting fantastic and daring deeds. Your complaints seem (from what I'm getting from your comments, I may be wrong) to fall into
This game does not simulate the minuata of skill difference that exists between individuals in a non-heroic setting
The game does not give the player a nice big ego enhancing number for being good at something
For the latter, well that's kinda a given. D&D is a epic heroic fantasy game. It's action, drama, and high stakes, high success. If you want a system that plays differently, play a different system.
For the latter, I don't know what I'd suggest to ameliorate that because it just seems like such a non-issue. It comes across like two people each cooking a meal and both getting equal praise for how delicious it is, but one getting mad because they should get more praise because they went to culinary school and practice cooking 5 hours a day, 7 days a week and have a michelin star.
While he is acknowledging that AI needs to be used responsibly the things he is talking about using it for are very troubling IMO. The statement is clearly not a 'we will use AI grammar checking for human generated story lines' but rather that AI will be in some way involved much more fundamentally in game development which IMO is a reversal of their prior stance and very quickly gets into exactly the bad stuff nobody wants. I don't think we will necessarily see AI images or entire books, they won't likely do that for copyright reasons at least right now but I would like to see much more of a firewall than what he is describing. If I want to consume material made by actual artists/writers not what an AI scraped off of actual humans and regurgitated. Sure some AI concepts might be fine but that is not something I am willing to pay for.
I agree with your interpretation of his words, given what's been provided. However, I'd also be wary of taking it to be meaningfully true. CEOs in my experience don't know what's going on in the factory floor of their own company. When I was working in one company, we had a massive new initiative that had a massive push from higher ups. The CEO came to visit, and had no clue what was going on. My manager and I had to brief him on it.
Cocks isn't even in Wizards, let alone be knowledgeable of and dictating the company's floor work. He can be as enthusiastic and passionate about AI as he wants, it probably doesn't make any difference to what Wizards are doing regarding its use.
I'm not going to tell you what you should think or feel about this. If you feel that strongly that even the slightest hint of it makes you want to boycott, then that's your prerogative - personally, I have greater qualms than AI but that's me. However, I'd consider waiting for something more direct from Wizards - John Hight or below. Cocks isn't an authority on Wizards, as odd as that sounds.
Rollback Post to RevisionRollBack
If you're not willing or able to to discuss in good faith, then don't be surprised if I don't respond, there are better things in life for me to do than humour you. This signature is that response.
I understand coming from 3.5 the numbers must seem small. Under that skill system, with piles of skill points and synergy bonuses a level 20 character will have a considerably higher number in a skill than a level 1 character. But in practice that just meant the DM had to increase the DC to make things a challenge. It was just a treadmill of every level, the characters got a +1 to everything, so the challenges also got a +1 to everything and you ended up in the same place as you started. 5e uses bounded accuracy to keep the numbers smaller and more manageable. It has the side benefit of not needing to create, say 7 different kinds of orc to allow orcs to be a challenge across multiple levels. It gets off the treadmill with fewer increases. It also means you can play the whole way through 1-20 without magic items, where past editions these were basically required to keep up with the monsters — just another treadmill.
When I started I remember thinking along those same lines, the number were so small, it was weird getting used to it. But I much prefer this version. You really get to the same place in terms of challenging characters, and the math is much tighter and easier to work with.
And to your point about the difference between an expert and a beginner, a level 20 character with expertise (which would reflect a true expert) and a max stat will have a +17 — higher if they have a generous DM with the right magic items. A level 1, with proficiency and not a good stat will be around a +3 or 4. That’s a very big difference, and I think gets at what you are talking about. Also there skill monkey classes and subclasses with a feature that basically says they can’t roll below an 10 on some skill checks, which really sets them apart.
And keep in mind, in this edition a 1 on a skill check is not an auto fail and a 20 is not an auto-succeed. So the level 20 character can’t get below an 18, and that level 1 character will never be able to succeed on a DC 25 check.
I’m kind of getting the impression you haven’t played a lot of 5e and that these complaints stem from looking at the differences and seeing smaller numbers. Maybe give it a shot for a while — it works well in play.
And, again, WotC keeps saying they won’t use ai and are very much against it. There have been a couple cases where a freelance artist used ai in the work they submitted, without the knowledge of WotC. Those artists have faced consequences. Will this always be their policy? Who knows, but it’s what we have to work with now.
Those are good points. The CEO of hasbro may not know what specifically is happening at wizards.
You still haven't stated what the CEO apparently said. If you could provide a link to this quote, others might be able to—at the very least—give informed comments, if not clarify what you might be misunderstanding.
This vague dancing around so supposed quote does not instil confidence
Edit: Correction and apologies, you have posted a link to the comment and I missed it. That's on me, sorry
Edit the second: these comments are relatively old news (literally over a year old) and states what has previously been mentioned in this thread—that WotC is using AI internally as a development tool and not a generative tool.
Inside of development, we've already been using AI. It's mostly machine-learning-based AI or proprietary AI as opposed to a ChatGPT approach. We will deploy it significantly and liberally internally as both a knowledge worker aid and as a development aid
It's important to note that machine learning and large language models may fall under the same umbrella of "AI", they're not the same thing. And propriety AI would mean their own learning models trained on their own data pools, not the plagiarized data pools of the likes of ChatGPT.
Also this is the in the same article:
We need to do it carefully, we need to do it responsibly, we need to make sure we pay creators for their work, and we need to make sure we're clear when something is AI-generated.
and
Wizards of the Coast representatives has repeatedly said that Dungeons & Dragons is a game made by people for people, as multiple AI controversies has surrounded the brand and its parent company. Wizards updated its freelance contracts to explicitly prohibit use of AI and has pulled down AI-generated artwork that was submitted for Bigby's Presents: Glory of the Giants in 2023 after they learned it was made using AI tools.
Those are good points. The CEO of hasbro may not know what specifically is happening at wizards.
You still haven't stated what the CEO apparently said. If you could provide a link to this quote, others might be able to—at the very least—give informed comments, if not clarify what you might be misunderstanding.
This vague dancing around so supposed quote does not instil confidence
They already have provided a link to it.
Rollback Post to RevisionRollBack
If you're not willing or able to to discuss in good faith, then don't be surprised if I don't respond, there are better things in life for me to do than humour you. This signature is that response.
I understand coming from 3.5 the numbers must seem small. Under that skill system, with piles of skill points and synergy bonuses a level 20 character will have a considerably higher number in a skill than a level 1 character.
...
I’m kind of getting the impression you haven’t played a lot of 5e and that these complaints stem from looking at the differences and seeing smaller numbers. Maybe give it a shot for a while — it works well in play.
That may be part of the issue but I see an underlying problem with the pared down skill list and binary skills. I don't have a particular issue with the numbers being smaller as much as the influence of degree of skill and the influence of natural ability still being very low. IMO those things should have a more equal weight with the dice and with circumstances that give you an advantage or disadvantage. And yes there are things like expertise and class features that smooth that out some. I am not a really big fan of that way of handling it but at least it does exist.
It may just be a byproduct if the epic fantasy nature of the game but at least to some extent it doesn't feel necessary outside of combat. Yes the characters get to a point with certain skills where they are gods among men if the skill and abilities count for more of the total roll... but that is very much in keeping with playing the heros in an epic fantasy setting. I have no issue with capping skills or stats in some way, but I think skill should count for more and have more variation than it does. In this addition there really is no way to represent on the character sheet having a bit of skill at cooking vs. that being a major focus of your character. I mean for cooking specifically there really isn't a way to indicate it at all. That's what I meant by not supporting out of combat stuff. In the combat there is a whole continuum of armor classes and the many many abilities while they can be very confusing for a new player do lead to a ton of variation in what different characters can accomplish etc. But outside combat there really are very limited mechanics to support differences between characters etc.
And, again, WotC keeps saying they won’t use ai and are very much against it. There have been a couple cases where a freelance artist used ai in the work they submitted, without the knowledge of WotC. Those artists have faced consequences. Will this always be their policy? Who knows, but it’s what we have to work with now.
The CEO of Hasbro seemed to be indicating it would not. But I am aware that was previously the policy and I was happy with that.
This game does not simulate the minuata of skill difference that exists between individuals in a non-heroic setting
The game does not give the player a nice big ego enhancing number for being good at something
Sort of but not quite. The issue isn't so much simulating the minutia of difference between skill levels but rather the degree to which skill vs. random chance affect outcome. I think skill and natural talent are more important than the system credits, compared to random chance, or modifying circumstances that give you advantage/disadvantage, which I also do not like the binary nature of though it is a solid mechanic for some things. But yes also ANY variation in level of skill beyond proficient not proficient. In the 3.5 system you could splash a couple points into cooking for flavor, or have a few points in a knowledge without it being a massive commitment for the character. IMO that was superior. I don't care if the number is big or small if there is a clear mechanical advantage to being good or not good at something. IMO the system doesn't mechanically reflect the differences between levels of skill or skilled vs unskilled people very well at all.
I don't care if the number is big or small if there is a clear mechanical advantage to being good or not good at something. IMO the system doesn't mechanically reflect the differences between levels of skill or skilled vs unskilled people very well at all.
This is somewhat of a design choice, not lack of roleplaying choices. It comes down to "do you want to have rolls that only one person in the group can realistically achieve, or do you want it to be something that everyone can join in on".
In the 3.5 system you could splash a couple points into cooking for flavor, or have a few points in a knowledge without it being a massive commitment for the character.
Here's the thing, "splashing a few points into cooking for flavor" is functionally identical to just saying "my character likes to cook". You don't need some mechanical notch on your sheet to flavour your character a certain way. This seems to be a mentality of "if there isn't a mechanical representation, it doesn't exist" which might explain your perception of the lack of roleplaying support if you're viewing it through that lens. And as for commitment issues with skills, skills are fairly easy to come by. You can gain skills through feats, backgrounds, training, bastion features etc. They're easy to come by to the point that there are discussions not just about how to make a build that gets all the skills, but what's the lowest level you can achieve that by. And skills give a solid mechanical reward that scales with level, meaning a small investment yields a high return.
I don't care if the number is big or small if there is a clear mechanical advantage to being good or not good at something. IMO the system doesn't mechanically reflect the differences between levels of skill or skilled vs unskilled people very well at all.
I think this is because you're still viewing it through the wrong lense, numerically speaking. The game features 3 tiers of skill level (not proficiency, proficiency, and expertise) combined with scaling based on level and ability score. Even two characters within the same "bracket" of proficiency can have differing levels of skill which directly translate into differing chances of success.
It should also be noted that the degrees of success percentage are not dissimilar from 3.5, it's just the numbers are different. It's just that a 5% step in success in 5th edition would be represented by a +1 whereas in 3.5 it would be represented by a +10
Degrees of success and failure exist within 5th edition and have done since 2014. Generally it takes the form of if you succeed or fail a check by 5 or more, there's an additional effect. There are also rules and suggestions for increments of 5. This is present in both the 2014 and 2024 DMG
I was always under the impression that degrees of success was more a 'homebrew' thing not in the official text. (For reference I use '14.) I'm aware of some degrees on saving throws like the medusa but elsewhere seemed to be more pass/fail. Where in the rules should I look for that kind of guidance in regards to skill checks etc.? (I'm genuinely curious.)
Degrees of success and failure exist within 5th edition and have done since 2014. Generally it takes the form of if you succeed or fail a check by 5 or more, there's an additional effect. There are also rules and suggestions for increments of 5. This is present in both the 2014 and 2024 DMG
I was always under the impression that degrees of success was more a 'homebrew' thing not in the official text. (For reference I use '14.) I'm aware of some degrees on saving throws like the medusa but elsewhere seemed to be more pass/fail. Where in the rules should I look for that kind of guidance in regards to skill checks etc.? (I'm genuinely curious.)
Honestly I picked it up from some creator but I believe this is the quote they were referencing. Not WOTC but Hasbro.
https://www.enworld.org/threads/hasbro-ceo-chris-cocks-talks-ai-usage-in-d-d-updated.706638/
While he is acknowledging that AI needs to be used responsibly the things he is talking about using it for are very troubling IMO. The statement is clearly not a 'we will use AI grammar checking for human generated story lines' but rather that AI will be in some way involved much more fundamentally in game development which IMO is a reversal of their prior stance and very quickly gets into exactly the bad stuff nobody wants. I don't think we will necessarily see AI images or entire books, they won't likely do that for copyright reasons at least right now but I would like to see much more of a firewall than what he is describing. If I want to consume material made by actual artists/writers not what an AI scraped off of actual humans and regurgitated. Sure some AI concepts might be fine but that is not something I am willing to pay for.
Reading this leads me to believe that you haven't actually looked into how D&D handles ability checks and skills as a system in general, so here's some info to expand your knowledge:
- Very Easy (DC 5) - 100%
- Easy (DC 10) - 85%
- Moderate (DC 15) - 55%
- Hard (DC 20) - 35%
- Very Hard (DC 25) - 10%
- Near Impossible (DC 30) - 0%
That's a very logical spread that isn't super random and that's for a level 1 character with no expertise. Let's compare this to a character making the same ability check with a non-key skill (no proficiency, +2 modifier)
- Very Easy (DC 5) - 90%
- Easy (DC 10) - 65%
- Moderate (DC 15) - 40%
- Hard (DC 20) - 25%
- Very Hard (DC 25) - 0%
- Near Impossible (DC 30) - 0%
Yes, there's an overlap of being able to do things, but that's intentional and (IMO good) design. Having it so only one character has a chance of doing something because they have the right skill is not great design in my opinion (something I touched on earlier). But if you think that's good design, the rules cater to that too because there is mention of only allowing characters to attempt ability checks if they have proficiency in the relevant skill, which would for example solve this below example:
There are the tools you want at your disposal in the rules. You've just got to actually look for them.
Oh, one more thing. I've googled the heck out of the matter and I can't find anything saying WotC is gonna start using AI to generate content. I've seen stuff about them using AI tools in their process for stuff like information management, database control, etc. And talk of building AI tools to assist DMs in forms such as rules engines and bots that can help you play. But zilch about generating content—everything has been very much "we're going to keep paying real humans to write the rules, make the art, and create the game"
Find my D&D Beyond articles here
Realityhackphotos, remember that ability scores are typically only made when there's a meaningful risk of failure or there's a time crunch. If a character with a +5 athletics bonus takes their time to carefully climb up a DC 15 wall, they will climb it successfully.
Yes that is true. And that papers over the problem in many situations. Unfortunately it doesn't fix them all. When you have an opposed roll, or a contest between two characters, or meaningful consequences for failure that doesn't fix the issue. But it does reduce how much you encounter it. It also doesn't really show the difference in what wall a highly skilled climber can climb carefully vs. what a novice can IMO.
To be honest I was in a bad mood when I made the OP and upset about the Hasbro CEOs comment and ranted rather harder than I intended to.
IMO 5th ed has some welcome improvements over 3.5 but they also went the wrong direction with some stuff, I understand it was largely a result of their attempt to simplify things but IMO they went too far and in the wrong direction entirely. Skills and stats should IMO have been made more meaningful not less, and the skill list expanded not trimmed. Yes there is complexity in that but it also opens up a lot of options for actively encouraging characters to be unique and fully reflect the players vision for them. I also think advantage / disadvantage is a fine mechanic but it is currently overemphasized at the expense of numerical modifiers that are sometimes just a better way to handle things.
These comments being?
Find my D&D Beyond articles here
I agree that *Some* overlap is good design I disagree with the amount of overlap. The difference between an untrained climber a trained climber and a top expert climber is a lot higher than what the system simulates. There absolutely are lots of things the untrained climber has zero shot at accomplishing that the expert climber can achieve reliably. And I don't have any problem with the game reflecting that. The issue with not allowing the skill check if they don't have proficiency it still does not allow for the wide range of abilities. Sure you can say nobody who is untrained can climb the wall, but that doesn't address the difference between a barely trained untallented person and one with extensive experience. There is expertise which provides a little bit of a patch for this issue but IMO it is inadequate.
More than that I also don't like the skills all being binary proficiency/non proficient. IMO this really limits the ability of the system to support a given characters traits. You cannot in the system simulate numerically being a little trained in x and extensively trained in y the way you could in previous editions. Outside of combat where there are a ton of different rules interacting to create much more variation in ability levels the system itself struggles IMO to support using rolling dice to determine outcomes.
ETA: And using advantage/disadvantage very heavily instead of numerical modifiers also creates very binary situations. You can't get extra advantage from additional favorable circumstances.
The comment in the linked article.
https://www.enworld.org/threads/hasbro-ceo-chris-cocks-talks-ai-usage-in-d-d-updated.706638/
This sounds like your objection lies in the numbers not massaging the ego. The game is perfectly set up to handle success and failure of epically heroic individuals attempting fantastic and daring deeds. Your complaints seem (from what I'm getting from your comments, I may be wrong) to fall into
For the latter, well that's kinda a given. D&D is a epic heroic fantasy game. It's action, drama, and high stakes, high success. If you want a system that plays differently, play a different system.
For the latter, I don't know what I'd suggest to ameliorate that because it just seems like such a non-issue. It comes across like two people each cooking a meal and both getting equal praise for how delicious it is, but one getting mad because they should get more praise because they went to culinary school and practice cooking 5 hours a day, 7 days a week and have a michelin star.
Find my D&D Beyond articles here
I agree with your interpretation of his words, given what's been provided. However, I'd also be wary of taking it to be meaningfully true. CEOs in my experience don't know what's going on in the factory floor of their own company. When I was working in one company, we had a massive new initiative that had a massive push from higher ups. The CEO came to visit, and had no clue what was going on. My manager and I had to brief him on it.
Cocks isn't even in Wizards, let alone be knowledgeable of and dictating the company's floor work. He can be as enthusiastic and passionate about AI as he wants, it probably doesn't make any difference to what Wizards are doing regarding its use.
I'm not going to tell you what you should think or feel about this. If you feel that strongly that even the slightest hint of it makes you want to boycott, then that's your prerogative - personally, I have greater qualms than AI but that's me. However, I'd consider waiting for something more direct from Wizards - John Hight or below. Cocks isn't an authority on Wizards, as odd as that sounds.
If you're not willing or able to to discuss in good faith, then don't be surprised if I don't respond, there are better things in life for me to do than humour you. This signature is that response.
I understand coming from 3.5 the numbers must seem small. Under that skill system, with piles of skill points and synergy bonuses a level 20 character will have a considerably higher number in a skill than a level 1 character.
But in practice that just meant the DM had to increase the DC to make things a challenge. It was just a treadmill of every level, the characters got a +1 to everything, so the challenges also got a +1 to everything and you ended up in the same place as you started.
5e uses bounded accuracy to keep the numbers smaller and more manageable. It has the side benefit of not needing to create, say 7 different kinds of orc to allow orcs to be a challenge across multiple levels. It gets off the treadmill with fewer increases. It also means you can play the whole way through 1-20 without magic items, where past editions these were basically required to keep up with the monsters — just another treadmill.
When I started I remember thinking along those same lines, the number were so small, it was weird getting used to it. But I much prefer this version. You really get to the same place in terms of challenging characters, and the math is much tighter and easier to work with.
And to your point about the difference between an expert and a beginner, a level 20 character with expertise (which would reflect a true expert) and a max stat will have a +17 — higher if they have a generous DM with the right magic items. A level 1, with proficiency and not a good stat will be around a +3 or 4. That’s a very big difference, and I think gets at what you are talking about. Also there skill monkey classes and subclasses with a feature that basically says they can’t roll below an 10 on some skill checks, which really sets them apart.
And keep in mind, in this edition a 1 on a skill check is not an auto fail and a 20 is not an auto-succeed. So the level 20 character can’t get below an 18, and that level 1 character will never be able to succeed on a DC 25 check.
I’m kind of getting the impression you haven’t played a lot of 5e and that these complaints stem from looking at the differences and seeing smaller numbers. Maybe give it a shot for a while — it works well in play.
And, again, WotC keeps saying they won’t use ai and are very much against it. There have been a couple cases where a freelance artist used ai in the work they submitted, without the knowledge of WotC. Those artists have faced consequences. Will this always be their policy? Who knows, but it’s what we have to work with now.
Those are good points. The CEO of hasbro may not know what specifically is happening at wizards.
You still haven't stated what the CEO apparently said. If you could provide a link to this quote, others might be able to—at the very least—give informed comments, if not clarify what you might be misunderstanding.This vague dancing around so supposed quote does not instil confidenceEdit: Correction and apologies, you have posted a link to the comment and I missed it. That's on me, sorry
Edit the second: these comments are relatively old news (literally over a year old) and states what has previously been mentioned in this thread—that WotC is using AI internally as a development tool and not a generative tool.
It's important to note that machine learning and large language models may fall under the same umbrella of "AI", they're not the same thing. And propriety AI would mean their own learning models trained on their own data pools, not the plagiarized data pools of the likes of ChatGPT.
Also this is the in the same article:
and
Find my D&D Beyond articles here
They already have provided a link to it.
If you're not willing or able to to discuss in good faith, then don't be surprised if I don't respond, there are better things in life for me to do than humour you. This signature is that response.
That may be part of the issue but I see an underlying problem with the pared down skill list and binary skills.
I don't have a particular issue with the numbers being smaller as much as the influence of degree of skill and the influence of natural ability still being very low. IMO those things should have a more equal weight with the dice and with circumstances that give you an advantage or disadvantage.
And yes there are things like expertise and class features that smooth that out some. I am not a really big fan of that way of handling it but at least it does exist.
It may just be a byproduct if the epic fantasy nature of the game but at least to some extent it doesn't feel necessary outside of combat. Yes the characters get to a point with certain skills where they are gods among men if the skill and abilities count for more of the total roll... but that is very much in keeping with playing the heros in an epic fantasy setting. I have no issue with capping skills or stats in some way, but I think skill should count for more and have more variation than it does. In this addition there really is no way to represent on the character sheet having a bit of skill at cooking vs. that being a major focus of your character. I mean for cooking specifically there really isn't a way to indicate it at all. That's what I meant by not supporting out of combat stuff. In the combat there is a whole continuum of armor classes and the many many abilities while they can be very confusing for a new player do lead to a ton of variation in what different characters can accomplish etc. But outside combat there really are very limited mechanics to support differences between characters etc.
The CEO of Hasbro seemed to be indicating it would not. But I am aware that was previously the policy and I was happy with that.
Sort of but not quite.
The issue isn't so much simulating the minutia of difference between skill levels but rather the degree to which skill vs. random chance affect outcome. I think skill and natural talent are more important than the system credits, compared to random chance, or modifying circumstances that give you advantage/disadvantage, which I also do not like the binary nature of though it is a solid mechanic for some things.
But yes also ANY variation in level of skill beyond proficient not proficient. In the 3.5 system you could splash a couple points into cooking for flavor, or have a few points in a knowledge without it being a massive commitment for the character. IMO that was superior.
I don't care if the number is big or small if there is a clear mechanical advantage to being good or not good at something. IMO the system doesn't mechanically reflect the differences between levels of skill or skilled vs unskilled people very well at all.
This is somewhat of a design choice, not lack of roleplaying choices. It comes down to "do you want to have rolls that only one person in the group can realistically achieve, or do you want it to be something that everyone can join in on".
Here's the thing, "splashing a few points into cooking for flavor" is functionally identical to just saying "my character likes to cook". You don't need some mechanical notch on your sheet to flavour your character a certain way. This seems to be a mentality of "if there isn't a mechanical representation, it doesn't exist" which might explain your perception of the lack of roleplaying support if you're viewing it through that lens. And as for commitment issues with skills, skills are fairly easy to come by. You can gain skills through feats, backgrounds, training, bastion features etc. They're easy to come by to the point that there are discussions not just about how to make a build that gets all the skills, but what's the lowest level you can achieve that by. And skills give a solid mechanical reward that scales with level, meaning a small investment yields a high return.
I think this is because you're still viewing it through the wrong lense, numerically speaking. The game features 3 tiers of skill level (not proficiency, proficiency, and expertise) combined with scaling based on level and ability score. Even two characters within the same "bracket" of proficiency can have differing levels of skill which directly translate into differing chances of success.
It should also be noted that the degrees of success percentage are not dissimilar from 3.5, it's just the numbers are different. It's just that a 5% step in success in 5th edition would be represented by a +1 whereas in 3.5 it would be represented by a +10
Find my D&D Beyond articles here
I was always under the impression that degrees of success was more a 'homebrew' thing not in the official text. (For reference I use '14.) I'm aware of some degrees on saving throws like the medusa but elsewhere seemed to be more pass/fail. Where in the rules should I look for that kind of guidance in regards to skill checks etc.? (I'm genuinely curious.)
This is a signature. It was a simple signature. But it has been upgraded.
Belolonandalogalo Malololomologalo Tumagalokumagalo, Sunny
Eggo Lass, Bone and Oblivion
Tendilius Mondhaven Paxaramus, Drakkenheim
Get rickrolled here. Awesome music here. Track 50, 9/23/25, The Mystery of Your Gift
https://www.dndbeyond.com/sources/dnd/dmg-2014/running-the-game#DegreesofFailure