Stability AI Rolls Out New 3D Model Tool

Stability AI, best known for its acclaimed AI image generator Stable Diffusion, just rolled out Stable 3D, a suite of 3D AI-based tools.

The new tool heralds a significant step forward, promising to make the creation of 3D models as simple as typing a description or uploading an image. Designed to cater to both seasoned professionals and those who are completely new to this new technology, Stable 3D is poised to make the intricate process of 3D modeling more accessible and efficient.

Getty Images v. Stability AI

As Stability AI strides into the future with Stable 3D, the shadow of legal confrontations looms over the company, given the ongoing copyright infringement lawsuits it faces from Getty Images and other artists.

In February, Getty Images filed a lawsuit against Stability AI, alleging that the company used its AI art generator, Stable Diffusion, to copy more than 12 million protected images from its database without permission or providing compensation using its AI art generator, Stable Diffusion.

The company is also named in another class action lawsuit, alongside AI startup Midjourney, by various artists who also allege copyright infringement.

Currently in private beta, Stable 3D aims to address the commonly time-intensive nature of 3D content creation, which for even simple assets, can take experienced creators hours or days to complete. By enabling the rapid generation of thousands of models a day, Stable 3D offers a potential boon for graphic designers, artists, and game developers by dramatically accelerating their workflow.

The company’s latest innovation allows users to transform natural language or 2D images into .obj format 3D models, which are compatible with mainstream 3D editing software such as Blender, Maya, Unreal Engine, and Unity. This capability could drastically lower the barrier to entry for creating 3D assets, providing a powerful tool for independent creators who previously may have been constrained by the need for great technical skills or resources.

However, in the wake of past contentions surrounding Stable Diffusion, the method by which Stable 3D’s generative models have been trained remains undisclosed, raising questions about the provenance and licensing of its data—especially in the context of the ongoing legal allegations against the company. 

Look at the Sky

As part of the Nov. 8 rollout, Sky Replacer is one of the more interesting features from Stability AI, which allows users to give life to their images – new and old – by swapping out the sky with one of nine dynamic alternatives like vibrant sunsets or a clear, blue canvas. 

Realtors, for example, can ensure that their property listings always appear in the most favorable light, without having to adjust photoshoots due to unfavorable weather conditions. 

Indeed, Stable 3D is entering a competitive field, one that has seen ventures from startups and tech behemoths alike. Yet, its connection with the widely adopted Stable Diffusion might provide an edge over contemporaries like Kaedim, Auctoria, and tech giants such as Meta and Nvidia.

The journey ahead for Stability AI is fraught with challenges as it seeks to position its Stable 3D at the forefront of AI-driven creativity while grappling with serious legal accusations. 

In August, a U.S. federal court upheld a previous decision made by the U.S. Copyright Office to rightfully deny copyright protection for any submitted generative AI works in the U.S. due to the failure of meeting the “authorship” requirement under the U.S. Copyright and Lanham Acts. 

The company’s strategy to maintain its market position includes diversifying its offerings, with its Stable Audio music generator, chatbot, photo editing features, and other generative tools.

However, amidst the excitement surrounding its technological prowess, Stability AI must also address its reported financial struggles and the complexities of translating its suite of AI tools into a sustainable business model—all under the watchful eyes of artists and industry watchdogs vigilant about protecting their copyrights.

Editor’s note: This article was written by an nft now staff member in collaboration with OpenAI’s GPT-4.