Google DeepMind is opening up entry to Project Genie, its AI instrument for creating interactive sport worlds from textual content prompts or photos.
Starting Thursday, Google AI Ultra subscribers in the U.S. can mess around with the experimental analysis prototype, which is powered by a mixture of Google’s newest world mannequin Genie 3, its image-generation mannequin Nano Banana Pro, and Gemini.
Coming 5 months after Genie 3’s analysis preview, the transfer is a part of a broader push to collect consumer suggestions and coaching knowledge as DeepMind races to develop extra succesful world fashions.
World fashions are AI techniques that generate an inside illustration of an surroundings, and can be utilized to foretell future outcomes and plan actions. Many AI leaders, together with these at DeepMind, consider world fashions are an important step to reaching synthetic basic intelligence (AGI). But in the nearer time period, labs like DeepMind envision a go-to-market plan that begins with video video games and different types of leisure and branches out into coaching embodied brokers (aka robots) in simulation.
DeepMind’s launch of Project Genie comes because the world mannequin race is starting to warmth up. Fei-Fei Li’s World Labs late final yr launched its first business product known as Marble. Runway, the AI video-generation startup, has additionally launched a world mannequin just lately. And former Meta chief scientist Yann LeCun’s startup AMI Labs may also concentrate on creating world fashions.
“I think it’s exciting to be in a place where we can have more people access it and give us feedback,” Shlomi Fruchter, a analysis director at DeepMind, advised TechCrunch through video interview, smiling ear-to-ear in clear pleasure over Project Genie’s launch.
DeepMind researchers that TechCrunch spoke to had been upfront concerning the instrument’s experimental nature. It might be inconsistent, typically impressively producing playable worlds, different occasions producing baffling outcomes that miss the mark. Here’s the way it works.
Techcrunch occasion
Boston, MA
|
June 23, 2026
A claymation-style fortress in the sky made from marshmallows and sweetImage Credits:TechCrunch
You begin with a “world sketch” by offering textual content prompts for each the surroundings and a essential character, whom you’ll later be capable of maneuver by means of the world in both first- or third-person view. Nano Banana Pro creates a picture based mostly on the prompts that you may, in concept, modify earlier than Genie makes use of the picture as a leaping off level for an interactive world. The modifications principally labored, however the mannequin often stumbled and would offer you purple hair whenever you requested for inexperienced.
You may use real-life photographs as a baseline for the mannequin to construct a world on, which, once more, was hit and miss. (More on that later.)
Once you’re glad with the picture, it takes just a few seconds for Project Genie to create an explorable world. You may remix current worlds into new interpretations by constructing on high of their prompts, or discover curated worlds in the gallery or through the randomizer instrument for inspiration. You can then obtain movies of the world you simply explored.
DeepMind is just granting 60 seconds of world era and navigation in the mean time, in half as a result of price range and compute constraints. Because Genie 3 is an auto-regressive mannequin, it takes plenty of devoted compute — which places a decent ceiling on how a lot DeepMind is ready to present to customers.
“The reason we limit it to 60 seconds is because we wanted to bring it to more users,” Fruchter mentioned. “Basically when you’re using it, there’s a chip somewhere that’s only yours and it’s being dedicated to your session.”
He added that extending it past 60 seconds would diminish the incremental worth of the testing.
“The environments are interesting, but at some point, because of their level of interaction the dynamism of the environment is somewhat limited. Still, we see that as a limitation we hope to improve on.”
Whimsy works, realism doesn’t
Google acquired a cease-and-desist from Disney final yr, so it wouldn’t construct fashions that had been Disney-relatedImage Credits:TechCrunch
When I used the mannequin, the security guardrails had been already up and operating. I couldn’t generate something resembling nudity, nor might I generate worlds that even remotely sniffed of Disney or different copyrighted materials. (In December, Disney hit Google with a cease-and-desist, accusing the agency’s AI fashions of copyright infringement by coaching on Disney’s characters and IP and producing unauthorized content material, amongst different issues.) I couldn’t even get Genie to generate worlds of mermaids exploring underwater fantasy lands or ice queens in their wintery castles.
Still, the demo was deeply spectacular. The first world I built was an try to dwell out a small childhood fantasy, in which I might discover a fortress in the clouds made up of marshmallows with a chocolate sauce river and bushes made from sweet. (Yes, I was a chubby child.) I requested the mannequin to do it in claymation fashion, and it delivered a whimsical world that childhood me would have eaten up; the fortress’s pastel-and-white coloured spires and turrets wanting puffy and engaging sufficient to tear off a bit and dunk into the chocolate moat. (Video above.)
A “Game of Thrones” impressed world that didn’t generate as photo-realistically as I neededImage Credits:TechCrunch
That mentioned, Project Genie nonetheless has some kinks to work out.
The fashions excelled at creating worlds based mostly on inventive prompts, like utilizing watercolors, anime fashion, or traditional cartoon aesthetics. But it tended to fail when it got here to photorealistic or cinematic worlds, typically popping out wanting like a online game slightly than actual folks in an actual setting.
It additionally didn’t all the time reply nicely when given actual photographs to work with. When I gave it a photograph of my workplace and requested it to create a world based mostly on the photograph precisely because it was, it gave me a world that had a few of the identical furnishings of my workplace — a picket desk, vegetation, a gray sofa — laid out in a different way. And it seemed sterile, digital, not lifelike.
When I fed it a photograph of my desk with a stuffed toy, Project Genie animated the toy navigating the area, and even had different objects often react because it moved previous them.
That interactivity is one thing DeepMind is engaged on enhancing. There had been a number of events when my characters walked proper by means of partitions or different stable objects.
I requested Project Genie to animate a stuffed toy (Bingo Bronson) so it might discover my deskImage Credits:TechCrunch
When DeepMind launched Genie 3 initially, researchers highlighted how the mannequin’s auto-regressive structure meant that it might keep in mind what it had generated, so I needed to check that by returning to elements of the surroundings it generated already to see if it might be the identical. For probably the most half, the mannequin succeeded. In one case, I generated a cat exploring yet one more desk, and solely as soon as when I turned again to the best aspect of the desk did the mannequin generate a second mug.
The half I discovered most irritating was the way in which you navigated the area utilizing the arrows to go searching, the spacebar to leap or ascend, and the W-A-S-D keys to maneuver. I’m not a gamer, so this didn’t come naturally to me, however the keys had been typically non-responsive, or they despatched you in the fallacious course. Trying to stroll from one aspect of the room to a doorway on the opposite aspect typically turned a chaotic zigzagging train, like attempting to steer a buying cart with a damaged wheel.
Fruchter assured me that his crew was conscious of those shortcomings, reminding me once more that Project Genie is an experimental prototype. In the long run, he mentioned, the crew hopes to boost the realism and enhance interplay capabilities, together with giving customers extra management over actions and environments.
“We don’t think about [Project Genie] as an end-to-end product that people can go back to everyday, but we think there is already a glimpse of something that’s interesting and unique and can’t be done in another way,” he mentioned.
