Skip to main content

On Sale: GamesAssetsToolsTabletopComics
Indie game storeFree gamesFun gamesHorror games
Game developmentAssetsComics
SalesBundles
Jobs
TagsGame Engines

From "Dream Logic" to Architectural Design: Where do we draw the line with generative models like Genie 4 ?

A topic by lapinchien created 5 days ago Views: 48 Replies: 3
Viewing posts 1 to 3
Host

Hi everyone,

Following up on the discussions around our Humans vs AI theme, I wanted to open a deeper debate on how we approach creation when tools like Google Genie 4 start generating environments and realities in real-time.

Some voices in our community have rightly expressed skepticism—pointing out the risk of falling into a chaotic "dream logic" where infinite generation replaces actual design, joking that it feels like building a house with no doors or windows and calling it avant-garde architecture.

The core question for this jam isn't whether we let the AI build everything, but who holds the blueprint. If the machine generates the corridors, the creator becomes the storyverse architect—the one who decides where the walls stand, where the doors open, and what the player is actually forced to confront.

I put together a short video exploring this shift from traditional asset-making to architectural curation with generative models:

How are you balancing emergent/generative elements with strict structural design in your entries for this jam? Are you leaning into the chaos, or tightening the walls?

Let’s discuss.

Host

I started this discussion on [itch.io Community » General » General Discussion]—feel free to join in! The exchange is genuinely interesting and inspiring, with open-minded, sharp participants who aren't trying to push fear-mongering agendas, but are truly eager to debate and deepen their own introspection.

/t/7248579/how-generative-world-models-like-googles-genie-will-re...

I don’t understand the question that much 😅. Can you be more specific? What do you mean with creating “environments”? I suppose it would mean to follow the player through a continuously improvised setting by the AI, following a certain story provided by the developer, but if the setting is improvised, wouldn’t the game be different for everyone? I don’t see that as much stable…

Host

Great point, olistupid! It’s completely natural to wonder if an AI-generated setting would just break down into pure randomness.

The distinction here is between unconstrained generation and modern World Models like Genie 4:

  1. Spatial Memory & Persistence: Unlike early generative tools that 'hallucinated' a new background frame-by-frame, models like Genie 4 maintain object permanence and scene geometry. If a player turns around and comes back to a room, the door, table, and layout remain identical.
  2. Conditioning on the Designer's Blueprint: The developer inputs strict seed parameters (rules, story triggers, layout bounds). The AI doesn't improvise what the world is; it handles the continuous real-time rendering of that defined world.
  3. Shared Seeds = Same Game for Everyone: Just like procedural generation in games like No Man's Sky or Minecraft, feeding the same seed/prompt configuration ensures that every player experiences the exact same core world architecture, while allowing small dynamic variations.

If you're curious about how Google addresses world stability, physical causation, and persistent visual memory in Genie 4, Google DeepMind published a detailed overview on their blog:

🔗 DeepMind - Introducing Genie 4: Our new world model for real-time interactive simulation