Stability AI, the company behind the text-to-image AI model Stable Diffusion, recently launched Stable 3D. This AI-powered application generates textured 3D objects, targeting non-experts and offering a streamlined process for creating 3D models. Despite the potential legal implications related to the data used in AI model training, this development marks a significant leap for Stability AI in the burgeoning field of AI-powered 3D model generation.
Stability AI, the innovative startup that brought us the text-to-image AI model Stable Diffusion, has sent a clear signal to the industry with the launch of Stable 3D. This AI-powered application aims to generate textured 3D objects for platforms used in modeling and game development, such as Blender, Maya, Unreal Engine, and Unity.
“Stable 3D: Empowering Independent Designers, Artists, and Developers with AI-Generated 3D Models.”
Streamlining 3D Model Creation
Stable 3D is currently available in a private preview for customers who have contacted Stability AI. According to the company’s blog, the tool is designed to help non-experts generate ‘draft-quality‘ 3D models swiftly. This development is a game-changer for graphic designers, digital artists, and game developers, as creating 3D content is often complex and time-consuming.
Stable 3D stands out for its robustness and capabilities compared to other model-generating tools on the market. Users can describe in natural language the 3D model they wish to create, or they can upload an existing image or illustration to convert into a model. Stable 3D outputs 3D models in the “.obj” file format, which can be edited and manipulated using most standard 3D modeling tools.
Addressing Potential Legal Ramifications
While this technological advancement is promising, it raises concerns about the data used to train Stable 3D. It is well-known that generative AI models tend to regurgitate training data. Consequently, if any data used was copyrighted and Stability AI did not obtain the proper licensing, Stable 3D customers could inadvertently incorporate IP-infringing work into their projects.
Stability AI has previously faced lawsuits for allegedly copying and processing millions of owned images to train Stable Diffusion without proper notification or compensation. The company has since partnered with startup Spawning to respect ‘opt-out’ requests from artists, but it remains to be seen if this agreement covers Stable 3D’s training data.
The Future of AI-Powered 3D Model Generation
Despite potential legal issues, Stable 3D signals Stability AI’s entry into the nascent but rapidly growing field of AI-powered 3D model generation. The market already features 3D object-creating platforms such as 3DFY and Scenario, as well as startups like Kaedim, Auctoria, Mirage, Luma, and Hypothetic. Even industry giants like Autodesk and Nvidia are exploring this space with apps like Get3D and ClipForge.
Stable 3D is seen as Stability AI’s latest bid to diversify its business in the face of increasing competition from other art-creating generative AI platforms, including Midjourney and OpenAI. Despite financial challenges, the company recently raised $25 million through a convertible note, taking its total funding to over $125 million.
Expanding the AI-Powered Product Suite
As part of its ongoing efforts to differentiate and increase sales, Stability AI has announced new features for its online AI-powered photo editing suite. These include a model fine-tuning feature and a ‘sky replacer’ tool. These additions join a growing list of AI-powered products from Stability AI, including the music-generating suite Stable Audio, the doodle-creating app Stable Doodle, and a ChatGPT-like chatbot.
In conclusion, while the launch of Stable 3D is a significant stride for Stability AI, its success will largely depend on how the company navigates potential legal issues and competition in the burgeoning field of AI-powered 3D model generation.