What Will Become the Dominant Game of the AI Era? Beyond the Block

Building a world ourselves, inviting others into it, and playing together: Minecraft is already one of the best-known pioneering examples of this style of play. Players can build and explore together in the same world. Minecraft’s official guide

While watching videos about making games with AI, I began thinking about what might come next.

Blocks are simple, understandable components that make creation manageable for people. But if AI joins the process as a creative assistant, perhaps we can handle greater complexity without losing that accessibility. If so, the building blocks may no longer need to be literal blocks.

I imagine three directions: a complete game made by an individual, a polished game made by a company, and a place where many people bring worlds they have created.

By “dominant,” I do not mean replacing every other kind of game. I mean becoming a large place where creators and players continue to gather over a long period.

I do not expect any one of these three directions to erase the others. But of the three, I suspect the next major current may come from an AI-expanded form of the third. That is the idea I was discussing with AI.

Games may expand in three directions

The first is an independent game made by an individual or a small team.

With AI assisting production, more people may be able to attempt projects that were previously beyond their reach. A complete game filled with one person’s favorite things, or one devoted to a very particular kind of play: I still expect a great deal from individual works like these.

The second is a highly polished game made by a company.

Experienced creators will use AI too. I do not think the ability to finish a work—its controls, presentation, story, and balance—will become unnecessary. If production becomes more efficient, competition between companies over overall quality may become even more intense.

The third is a game in which individuals bring worlds and experiences onto a common foundation provided by a company.

A company supplies the underlying systems for movement, combat, saving, and multiplayer interaction. Individuals then use AI to create their own towns, dungeons, characters, and quests on top of that foundation. Players visit places made by many different creators.

This third style of play already exists. My prediction is that AI will broaden both the number of people who can participate and the range of things they can make.

This is not meant as a strict classification. Small companies exist, and companies can create works inside shared platforms. Of these broad directions, however, the third is the one that interests me most.

Thinking of Minecraft as something like LEGO

The starting question was: why could Minecraft become an accessible entrance into player creation?

I think one part of its appeal is that it simplifies the creative system so that individuals can begin making things.

With LEGO, understandable pieces can be combined into something a person has imagined. Children can begin building, while adults can take the same system much further. Minecraft seems to have a similar combination of accessibility and freedom.

This is my own interpretation of one reason for its popularity. I am not saying Minecraft itself lacks complexity. I am focusing on how understandable it is when someone begins to create.

As people grow, their interests may expand from block play to more complex toys and forms of creation.

Inside a game, after building a structure, a creator may want people to live there. They may want those people to talk, for incidents to occur in the town, and for adventures to begin. Eventually, they may want to create not only the appearance of a place but also what happens there.

Translating those ideas into code and detailed settings is difficult. There is a gap between what someone can imagine and what they can implement.

If AI helps, must the components still be blocks?

That led me to another thought.

One way to make creation accessible is to reduce what people build to simple components. Another may be to let AI assist with the complicated work.

A person decides what kind of town they want, who lives there, and what happens there. AI helps turn those ideas into something that operates on the shared platform.

If that becomes possible, we may no longer need to limit the complexity of what can be expressed to whatever the creator can manipulate directly.

What if someone could create a world containing characters, conversations, music, and events with an entrance as approachable as placing blocks? That is the center of my idea.

Existing creative spaces and their expansion through AI

Existing examples also matter here. I checked the following official information as of September 18, 2026; some of these examples were introduced to me during the conversation with AI.

Game or creation platform	What already exists
Minecraft	A game in which players build worlds from blocks. Creative mode provides unlimited resources for building. Official guide
Super Mario Maker 2	Players make courses from provided parts, upload them, and play courses made by others. Official Nintendo site
Roblox	A common platform on which experiences can be created and published with Roblox Studio. It also incorporates generative AI features that assist creation. Official developer documentation
Fortnite / UEFN	Tools for creating games and islands for Fortnite, using foundations such as multiplayer and combat while building original environments and play. Official Epic site

Roblox in particular announced a beta of “4D generation” in February 2026, generating three-dimensional objects with behaviors. In compatible experiences prepared by developers, the announcement demonstrated players generating objects such as drivable cars from text prompts. At the time of the announcement, this was a limited system, not a completed ability to generate every kind of world freely. Official Roblox announcement

In other words, common foundations for individual creation already exist, as does the movement toward AI-assisted production.

Creation on these platforms already goes far beyond placing blocks. What interests me is how far AI can lower the difficulty of taking part in that complex creation.

Could people without specialized implementation knowledge gain a broader entrance into making a “small world” containing characters, conversations, music, and events? The examples above do not mean that this entire idea has already been realized.

A wide horizon filled with small worlds

Visit a town made by one person and you find characters they imagined, conversations, and music. Visit a dungeon made by someone else and different enemies and devices await you.

Perhaps I use a cute character, then travel into a fully developed horror world made by someone else. The collision between different creators’ tastes could be interesting in itself.

During the conversation, I described this as “a wide horizon filled with small worlds.”

No one person has to complete an enormous game in its entirety. Each can contribute the place or experience they want to make. Players can visit worlds imagined by many people on the same foundation.

The third direction is where I see the greatest possibility

I think interesting works will continue to emerge both from individual games and from polished corporate productions.

Even so, as a place where people gather over time, I see particular potential in an AI-expanded version of the third direction. New play can be introduced not only by the provider of the foundation but also by the many creators who participate in it.

After finishing one world, a player travels to another person’s world. Someone who began as a player eventually starts creating too. A place that develops this kind of cycle could become very strong.

Of course, as freedom grows, maintaining common rules and helping people find worthwhile worlds become serious problems. Ease of creation alone does not make a place somewhere people want to play.

So I cannot say which company will succeed. But when I imagine a dominant game of the AI era, what comes to mind is a place where people can make and share more complex worlds and forms of play with an entrance as approachable as assembling blocks.

Minecraft showed the pleasure of making something and playing together. How far can AI extend that accessibility?

An individual game, a corporate game, and a place where people bring their own worlds: in a future where all three expand alongside one another, I want to see where the third direction leads.

---

The exchange that prompted this article

The following is excerpted from a conversation on September 18, 2026. Parts of the explanation have been omitted.

Me

But once we enter the AI era, it will not necessarily have to be blocks, will it? As people grow, they move from LEGO blocks to more complex toys. I think people used blocks because programming and managing fine details were difficult, so the smallest manageable unit of information was a block, and that format let everyone share and enjoy creating together.

AI

One especially strong part of your hypothesis is that you see Minecraft’s success not as a result of “blocks themselves,” but as the result of compressing the unit of creation down to a level of complexity people can manage.

If AI enters between the person and the system, the upper limit of manageable complexity rises dramatically.

The next generation of “LEGO blocks” therefore may not be square blocks at all. It may be the game systems themselves.

This article organizes my own view and an idea about the future through dialogue with AI. The AI response is quoted as the view expressed at the time. AI also assisted with the structure and writing.