eWEEK content material and product suggestions are editorially impartial. We might generate income while you click on on hyperlinks to our companions. Learn More.
London-based synthetic intelligence lab Odyssey has launched a public analysis preview of what it calls an interactive video — a brand new storytelling medium generated fully by AI in actual time.
Unlike conventional video or gaming experiences, Odyssey’s new system allows customers to observe, work together, and immediately affect the video because it performs, with only a keystroke, swipe, or joystick motion as enter.
Powered by a brand new world mannequin, the system can generate and stream video frames in response to person enter, making a seamless phantasm of interacting inside a reside digital world. Odyssey mentioned this expertise might mark the start of a serious shift in how video is created, consumed, and skilled.
A brand new medium for creatives
Founded by self-driving automobile veterans Oliver Cameron and Jeff Hawke, Odyssey started constructing world fashions to streamline movie and sport manufacturing; nevertheless, via their analysis, they found a completely new method to storytelling.
The AI lab mentioned this innovation isn’t meant to interchange creators however to empower them.
“Interactive video . . . opens the door to entirely new forms of entertainment, where stories can be generated and explored on demand, free from the constraints and costs of traditional production,” Odyssey wrote of their weblog.
A piece in progress
The startup acknowledged the expertise is much from excellent, noting that environments can change unexpectedly and visuals could also be blurry or distorted. “The experience today feels like exploring a glitchy dream — raw, unstable, but undeniably new,” the corporate mentioned.
Odyssey famous that, whereas the preview may be thought of a “narrow distribution model,” that means it’s educated on a particular set of real-world movies to keep up stability, their group is already engaged on a next-generation mannequin that guarantees sharper visuals, higher physics, and extra reasonable interactions.
The early demo can stream as much as 30 frames per second utilizing highly effective NVIDIA H100 GPUs, reacting to customers’ actions inside 40 milliseconds.
Odyssey’s huge goals
Odyssey has raised $27 million from traders, which embrace EQT Ventures, GV, and Air Street Capital. The startup’s board contains Ed Catmull, cofounder of Pixar and former president of Walt Disney Animation Studios.
The infrastructure value to run this AI instrument at the moment stands at $1 to $2 per user-hour. But Odyssey says, “This cost is decreasing fast, driven by model optimization, infrastructure investments, and tailwinds from language models.”
Zooming out, Odyssey sees a future the place every part video touches evolves into interactive experiences. “Over time, we believe everything that is video today — entertainment, ads, education, training, travel, and more — will evolve into interactive video,” the corporate wrote.
And the Odyssey group will not be alone. Other firms, together with Google DeepMind, Microsoft, and Fei-Fei Li’s World Labs, are additionally racing to develop world fashions able to constructing reasonable, responsive digital environments.
