We're not going to see a new game. Not unless Yoshi-P decides he's had enough of us. The result will be the game will be shutdown and that's it. No more MMO's.
The reason is, the market is absolutely diluted with low-quality, Unity or Unreal engine-based, cookie-cutter games that you can't barely tell apart. All gunning for your wallet with microtransactions.
Designing and reusing their own game engine both screwed and saved FFXIV. It screwed the game in V1.0 because they tried to adapt a single player game engine (the same happens with Unreal engine games) to a MMO, when the PC specs for the average PC player were only capable of running PS2 quality games. However those high quality efforts (other than the general copy-pastaness of the overworld and reuse in dungeons) also made the game very expensive to just throw away. FFXIV started as a potential sequel to import characters from XI and thus it retains races and lore bits that would have allowed for that.
But, WoW came out and the market decided that casual ramshackle lore easy mode was the direction everyone was running for, so players abandoned XIV V1.00 to play other games that only lasted 3-12 months before shutting down. Ironically V1.00 outlasted many other attempts to copy it's visual look. V2.00 was certainly a downgrade, visually, and for animations, and zone sizes. But it made it a different game. We've also been shown both the technical limits of this engine (FFXV) and what it can't do (Kingdom Hearts III, and thus was switched to Unreal.)
Now at present, the market is rapidly switching to plotless F2P microtransaction-filled MMOShooter Battle Royale's, which will probably have a lifespan of 5 years before the next thing comes along. So if SE decided to release it's own MMO, from scratch, in this time frame, that market will be gone by the time it's out.
SE's best bet, is to continue to update and upgrade FFXIV until it makes no business sense to. Square's traditionally been very good about storylines and only started to stumble post Square-Enix merger. If they've started on another MMO, here's the wishlist of stuff that I would want that would make me happily switch from FFXIV:
1. I will not play a VRMMO. The technology is at present rubbish, and until safe non-invasive BCI's are invented I think this is the wrong tree to bark up. I would also not play a MMO that looks photo-realistic. Sorry the uncanny valley called from 2001 (The Spirits Within) and said "this is expensive and the audience won't appreciate it". I think the visual style of FFXIV actually looks about right and prefer a clean scifi anime aesthetic. Where we can improve with today's technology is by pointing a camera and microphone at the player and having the game translate the motions and voices into their character's motions and generated voice as a secondary way to communicate with players.
2. Adaptive and forgiving network code. There is no reason players should have delayed commands, rubber banding or 90k errors. Most MMO's absolutely suck for players who don't live within a few miles of the data center, and always suck for players on wireless connections. Design around. To this end, consider rendering the game in the data center and streaming the output so everyone has the same experience. Then you can literately charge more for renting from the 4K/8K render center instead of the HD center. Plus it eliminates the ability to cheat or datamine the game. If a player's connection becomes rubbish, their session is still running on the render system until specifically logged out.
3. A Seamless openworld with sandbox qualities (eg destructible environments, build your own house or prefab from a template and have it persist throughout the game's lifecycle unless neglected.) Players could literately create and manage their own cities if they can stop fighting with each other long enough to do so. You can come back a year later and it will still be there if you left the resources to repair it inside.
4. Instance-free world. Except where the population would make parts of the game unplayable, all world-wide players should be able to be in the same world at the same time. Sectorize the world into smaller and smaller parts to support this rather than trying to create instances.
5. Closed-loop economy, closed-cycle belligerents. The game does not generate infinite resources, you have to find them, and once found, they are gone. Enemies, monsters, animals are both resources and creatures who will fight each other if a resource becomes hard to find. Harvest all the trees, then the birds will attack players over the few remaining. Kill all the cows, milk shortage. Players decide what is actually valuable and discover and trade those. Resources expire if not used, and spoiled resources have to be found again to prevent hording. Even equipment and housing will decay if not maintained. Certain resources are pseudo-unlimited (eg water, soil, sand, clay, rock) so players can make bricks, glass and rock structures out of materials that just exist, as they exist in the field in large quantities.
6. Once created, you never actually logout. You can disconnect from your session, but your character is always in the game. They will be given a schedule/job to do when you're not connected. When connected, you can take over where they were in their schedule and go adventuring. When logged out, they will be set to jobs like "Guard" (for combat skills, also involved in tax collection), craft Merchant (will produce or repair equipment on demand for players if materials available), Politician (for players who are involved in how the city grows, and rules imposed), Banker (if you've built a vault), Gardener/Farmer, and so forth. Basically players who are not connected to their session = NPC's. Disconnected players without a job will play it safe and just go to sleep in their designated house. No house, player will go find an Inn if they have money, or will sleep outside if they have a sleeping bag.
7. Layered lore. The problem with a lot of games right now is that they just run head first into a lore gate, and players make no mark upon the world. Sure "you" might be the WoL in this game, but every player is a WoL from their point of view, nothing they do impacts your story. Rather let's create lore that is more about working together to a common goal. Hence players vs the environment and players vs players (politically) should be the guiding narrative rather than adapting a single player narrative. If a story calls for the merchants guild to come together to solve a water shortage, then all the players who are merchants must be involved, and those who don't want to be part of the story can literately go "no thanks" and they will be ejected from their merchant job until that storyline runs its course. What starts that water shortage can be a repeatable bit of story that happens when players neglect maintenance of the city itself. Combat oriented lore is also dynamically engaged when certain populations of animals, monsters, or even jobless players thresholds are met. For example if there are more homeless players than city guards, then the cost to the politicians is that taxes must go up to hire more guards, or if there is an insufficient number of guards, PvP become enabled in areas not patrolled by guards, and damage to property becomes possible. The guiding story for the world could be classic Final Fantasy of trying to save the world from destruction, but it requires the players to meet goals to progress rather than just defeating some big-bad and then waiting 6 months for more content. Sure, there will be combat "big bads" that are players (a career could be literately be "super-villain") and "big bads" that are AI controlled monster teams, but they will also work together to meet their own goals. Sometimes it may even result in players being aligned with villains to save the world.
8. No finite end. The game should be able to persist even without the developers producing new content. This would require AI to actually determine what story bits should be played out when conditions are ripe for it. So some storyline may be submarined for a very long time since data miners would not be able to find it, even if they were staff at the publisher. Players can also make their own AI storylines when conditions are hit, even if they're not present when the conditions activate, and can then defer working on them until they connect.
9. If the player never logs back in, their character will continue to persist as long as they've been given some instruction to do something while not connected. Even if their subscription ends. Even if they pass away. I doubt most companies would be willing to run a game for decades, but that could be the thing that makes humans want "AI" to demand a right to exist.
Other than the last point where I was being a bit facetious, I don't believe any game publisher wants to put in the money to create a persistent world that is oriented around trying to solve humanities problems rather than some cheap microtransaction pew-pew-pew money grab. I think the above would likely have to come out of a deliberate AI research project that is adapted for gaming rather than the other way around.
But overall I think that gamers have grown up and don't want to just login to a repetitive grind. If you've been on the forums for any amount of time you will know that there are players who don't care one bit for the storyline and there are players who care more about the storyline than any of the combat that it involves. There is no way to please both audiences. What I propose above is meant to appease the former more than the latter, as the storyline is really optional by opting not to participate as the story unfolds.
tl;dnr version: I want a simulated world that has the story depth of a FF game, but not the grind of one.