AI Model: Revolutionary Tech Streams Interactive 3D Worlds
4 min read
BitcoinWorld AI Model: Revolutionary Tech Streams Interactive 3D Worlds Imagine stepping into a video, exploring its environment just like you would in a video game. This isn’t science fiction anymore. Odyssey, a startup co-founded by self-driving tech veterans Oliver Cameron and Jeff Hawke, is making this a reality with their new AI model designed for streaming interactive video. What Does This Revolutionary AI Model Do? At its core, Odyssey’s technology allows users to ‘interact’ with what looks like streaming video. Available via an early web demo, the AI model rapidly generates and streams video frames, pushing out a new frame every 40 milliseconds. This rapid generation enables viewers to navigate and explore areas within the video environment using basic controls, offering an experience akin to navigating a 3D-rendered video game rather than passively watching. Odyssey describes the underlying mechanism: “Given the current state of the world, an incoming action, and a history of states and actions, the model attempts to predict the next state of the world.” This predictive capability is key to creating a sense of continuity and responsiveness as the user moves through the generated environment. Exploring the Core Technology: The World Model Powering this interactive experience is a sophisticated world model . Odyssey claims this model demonstrates several capabilities crucial for creating believable, explorable environments: Generating pixels that appear realistic. Maintaining spatial consistency within the generated space. Learning actions directly from video data. Outputting coherent video streams for durations of five minutes or more. The development of advanced world model s is a significant area of research in the AI field, attracting attention from major players like DeepMind, Fei-Fei Lee’s World Labs, Microsoft, and Decart. The belief is that these models could one day underpin highly realistic simulations, interactive media, and even advanced training environments for robots. The Promise of Interactive Video Odyssey refers to its creation as “ interactive video .” The company envisions this technology fundamentally changing how we consume video content. Their blog post suggests that 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.” Looking ahead, Odyssey believes that “everything that is video today — entertainment, ads, education, training, travel, and more — will evolve into interactive video , all powered by Odyssey.” This represents a significant shift from passive consumption to active exploration within digital content. Navigating Challenges and Creative Concerns While the potential is exciting, the technology is still in its early stages. Odyssey acknowledges that their current demo is “a bit rough around the edges.” Users might encounter environments that are blurry, distorted, or even spatially inconsistent, where layouts can change unexpectedly as you move around. These are significant challenges in maintaining immersion and usability. Beyond the technical hurdles, the rise of AI in creative fields, including technologies like Odyssey’s, has sparked concerns among creative professionals. Investigations have highlighted how some companies are leveraging AI in ways that impact jobs, such as using AI to cut corners or address workforce attrition, as seen with game studios like Activision Blizzard. A 2024 study commissioned by the Animation Guild estimated that over 100,000 U.S.-based film, television, and animation jobs could be disrupted by AI in the near future. Odyssey has publicly stated its intention to collaborate with creative professionals rather than seeking to replace them, aiming to provide new tools and possibilities. Building the 3D Worlds : Technology and Approach Despite the demo’s current limitations, Odyssey is focused on rapid improvement. The model can stream video at up to 30 frames per second, powered by clusters of Nvidia H100 GPUs. The cost for this early tech is estimated at $1-$2 per “user-hour.” Odyssey is taking a unique approach to training its world model . Unlike many labs that rely solely on publicly available data, Odyssey has developed a specialized 360-degree, backpack-mounted camera system to capture real-world landscapes. They believe this proprietary data will enable the creation of higher-quality models and more realistic 3D worlds . The company is actively researching ways to improve the models, focusing on “richer world representations that capture dynamics far more faithfully, while increasing temporal stability and persistent state.” They are also working on expanding the range of user interactions beyond simple motion. With $27 million in funding from investors like EQT Ventures, GV, and Air Street Capital, and with industry figures like Ed Catmull (Pixar co-founder) on their board, Odyssey is positioned to push the boundaries of this technology. They are also developing software to allow creators to export scenes generated by their models into standard tools like Unreal Engine, Blender, and Adobe After Effects for further refinement and hand-editing, reinforcing their stated goal of collaboration. Odyssey AI: A Glimpse into the Future Odyssey’s work on streaming, interactive 3D worlds powered by a novel AI model represents a significant step towards a new era of digital content. While challenges remain and the impact on creative industries requires careful consideration and collaboration, the vision of a future where video is no longer a passive medium but an explorable, interactive experience is compelling. As Odyssey AI continues to refine its technology, the possibilities for entertainment, education, and simulation appear vast. To learn more about the latest AI model trends, explore our article on key developments shaping AI features. This post AI Model: Revolutionary Tech Streams Interactive 3D Worlds first appeared on BitcoinWorld and is written by Editorial Team

Source: Bitcoin World