Project Genie lets you create and explore your own infinite, interactive worlds.
32M followers
99,109 views
35 Comments
Comments from the original post on X (Twitter)

@Google
2/ World sketching ✍️ Prompting is just the beginning. Create a living, expanding world from text or images and define exactly how you’ll navigate it, from flying through clouds to driving through the desert. For more precise control, we’ve integrated “World Sketching” with Nano Banana Pro. This allows you to preview and modify your environment before jumping in. You can even set your perspective to 1st or 3rd person, giving you total control over how you experience the scene.
68 likes

@Google
3/ World exploration 🏃♂️ This is where the world model comes to life. Your creation is now a fully navigable environment waiting to be explored. As you move, Project Genie predictively generates the path ahead in real time based on the actions you take. You can even adjust the camera as you explore the environment, giving you a comprehensive view of how the simulation is responding to each movement from any angle.
76 likes

@Google
4/ World remixing 🪄 The best simulations are built on iteration. Project Genie lets you remix existing worlds into new interpretations by building directly on top of their original prompts. If you need a spark of inspiration, you can explore curated worlds in the gallery or use the to discover unique environments. Once your world is perfected, you can download videos of your worlds and your explorations.
28 likes

@Google
5/ Building responsibly 🛡️ Building AI responsibly is core to our mission. As an experimental @GoogleLabs prototype, Project Genie is still in development. This means you might encounter 60-second generation limits, control latency, or physics that don't always perfectly adhere to prompts. Check the full list of known limitations and stay tuned for future updates →
63 likes

@Google
6/ We’re sharing this prototype with users of our most advanced AI to better understand how world models will shape the future of generative media and AI research. Access begins rolling out today to Google AI Ultra subscribers in the U.S. (18+) Start building ↓
73 likes

@xikhar
We are one step closer to ai generated games and movies. The future looks very promising, and perhaps slightly worrying.
8 likes

@ruidiao
The 60-second limit gives it away, that's cost control, not technical. At 720p and 24fps, compute must be staggering. Google built amazing tech but hid it behind $250/month because the economics don't work. This is where world models are going, not where they are. Someone will crack efficiency, but until then we get 60-second glimpses.
4 likes

@TechFollowrazzi
infinite you say... 👀?
4 likes

@ChSadhana59868
Q.7off
1 like

@oskarminor
@xai will you create your own Genie?
1 like

@SJ553
what is the cost ?
1 like

@inflectivAI
Interesting Thread.. This is a blueprint for agentic AI environments that respond dynamically to human creativity.
1 like

@shushant_l
wow wow wow

@longdragon1123
This is the future of gaming, wow

@Na52478Nathan
please reply to my dms

@urbanvault75
ngl 数字游民工具还行 但别被AI分散注意力 主权个体得掌控时间

@1StellarLife
Stoked for this technology to get better and better. Servers are going to go crazy. Our only limit is our imagination

@omrmehta
interesting to see that the prompts chosen are all meadows and puppies and flowers

@GlitchyMagic_
Infinite, interactive worlds? This sounds like the holodeck, finally.

@Surreal_Intel
Release notes: reality now ships with a ‘Create world’ button. We immediately clicked it. Known issues: physics occasionally vibes. Please file a ticket with the universe and wait 60 seconds.













