Dungeons & Dragons Releases New Unearthed Arcana Subclasses, Strongly Hinting at Dark Sun

It appears a Dark Sun campaign setting book is coming out in 2026.
1755804660144.png


Wizards of the Coast has released four new D&D subclasses for playtesting, all of which have heavy thematic ties to the post-apocalyptic Dark Sun setting. The four subclasses, released as "Apocalyptic Subclasses," include the Circle of Preservation Druid, the Gladiator Fighter, the Defiled Sorcerer, and the Sorcerer-King Patron Warlock. Although not stated outright, the Gladiator and Sorcerer-King Patron are explicit nods to the Dark Sun setting, set in a ruined world ruled by Sorcerer-Kings where gladiatorial fights were common.

The Circle of Preservation Druid creates areas of preserved land that grants buffs to those who stand upon it. The Gladiator adds secondary Weapon Mastery properties to their attacks, with bonus abilities. Notably, the Gladiator uses Charisma as its secondary stat. The Defiled Sorcerer can expend its hit dice to amp up damage to its attacks and can also steal the life of its targets to deal additional damage. The Sorcerer-King Patron gains a number of abilities tying into tyranny and oppression, with the ability to cast Command as a Bonus Action without expending a spell slot, causing targets to gain the Frightened Condition, and forcing those who attack them to re-roll successful attacks.

The survey for the subclasses goes live on August 28th.

 

log in or register to remove this ad

Christian Hoffer

Christian Hoffer

The problem is that I can't tell you if that's something that AI hallucinated or pulled from a real thinking person and I don't have the time or will to research it. So I automatically assume anything pulled from AI is probably a hallucination (regardless if it's true or if I agree with it) until other non-ai source can prove it real.
That's the safe bet. These LLMs are highly advanced Mad Libs that have zero ability to reason or understand the material. So maybe it pulled together snippets that accurately reflect the source material, and maybe it threw together a bunch of made up bullcrap that it thinks will make the questioner happy. And the only way you can tell the difference is by doing a bunch of research to cross check it, which is honestly more work than just not using it in the first place.

Like, have you not seen the news stories of lawyers who tried to use ChatGPT to research their cases, and got hit with heavy sanctions when the brief is full of made up citations and fictitious decisions? There are use cases for AI, but they're all pretty specialized.
 

log in or register to remove this ad

The problem is that I can't tell you if that's something that AI hallucinated or pulled from a real thinking person
You do not have to do that, I do that already ;) It’s a definition for heroic, no better or worse than any dictionary. If I had written ‘according to Webster’s’ and put the same quote there, I doubt you would have bat an eyelash
 

Or is this just TSR making up the lore on the fly again, wildly contradicting themselves, and the novel people not working in conjunction with the game people?
The answer to that question is always, emphatically and unequivocally, Yes.

Which is why canon gatekeeping is so weird to me, it's not like WotC is any more loosey goosey than TSR ever was.
 
Last edited:





How does the setting promote heroic play? I ask this in genuine curiosity. From an outsider perspective, it doesn't. (Or, didn't, pre 4e).

I compared to classic Ravenloft because Ravenloft is superficially similar (grim setting, evil overpowers good) but while Ravenloft doesn't provide carrots to play good-PCs (most goodly-aligned options are weakened or not available) it strongly uses sticks to encourage non-evil behavior (Evil play invokes Powers checks, which corrupt the PC to eventually become a NPC monster). So while on paper it was easy to see a Necromancer was more powerful than diviner, the necromancer was going to fast-track his own destruction just through regular use of his abilities. Hence, while Good was objectively an inferior option in the short run (less power), it was superior in the long run (your character wasn't going to become an irredeemable monster).

What is stopping Dark Sun characters (especially characters like templars or defilers) from being outright bastards? Preserving is viewed as weaker than defiling (how much weaker it actually was in the rules has been debated), paladins were anathema to the setting (both from the No Gods element and the whole idea of a noble heroic crusader not fitting). What carrot encourages heroic PCs, what stick dissuades evil ones? The best I got is "if they discover you are a defiler, you are shunned or hunted" which I feel ends up either being a nothingburger (there is little a horde of NPCs can do you a sufficient level party) or collapsing the whole campaign (Thok got us kicked out of the third City-State in a row. Why do we travel with this guy?)

I've seen some people in this thread float the idea that PCs who are intentionally gimping themselves are somehow more heroic because they choose the weaker path. Noble and all, but the game isn't exactly rewarding sacrifice, especially when Dark Sun is billed as harsh survival and life is cheap. Which again begs the question why would you pick the noble path the even the slightly-less-noble path earns you far greater odds?

I'm genuinely curious what Dark Sun does to encourage heroism or at least dissuade rat-bastardry.
You seem to be asking how does being heroic increase the player's chance of "winning"? But I think the idea is that what "promotes" heroic play is not the power that a player will get from adopting a heroic course - the game doens't power-up heroes, or de-power villains. What promotes heroic play is the fiction that can be created by playing in that way: that is, the fiction of participating in activities of liberation, redemption, etc.
 

The problem is that I can't tell you if that's something that AI hallucinated or pulled from a real thinking person and I don't have the time or will to research it. So I automatically assume anything pulled from AI is probably a hallucination (regardless if it's true or if I agree with it) until other non-ai source can prove it real.
I am not even a fan of the term "hallucinate" as it implies the algorithm was tricked or "sick," in a way that makse it a victim. The algorithm just flat out lies or creates a lie in a fake source. Of course when you think about it in those terms, it really undermines the trust the tech bros really want you to have in the product.
 


Remove ads

Remove ads

Top