Sketchfab updates terms and conditions to protect users’ 3D models from AI machine learning

Sketchfab – one of the largest platforms with 3D models, has changed its terms of use, which is directly related to recent changes in tools based on AI (artificial intelligence). From now on, users of the platform uploading their work to it can select the “NoAI” option, which will mean that their 3D models will not be made available under the standard Creative Commons license and will not be able to be used by AI tool developers to use them in the machine learning process.

For several months, thousands of people around the world have been using tools based on artificial intelligence (AI) to generate content, images or (recently) music. Tools such as ChatGPT, DALL-E or Midjourney are able to generate any content or images based on a text description. The more precise and detailed the task description is, the better the end result. Although the above tools were made available to users only a few months ago, they are already the source of numerous controversies related to copyright issues.

Because the source of all content and works generated by AI are millions of texts, images or photos that have been processed by AI systems during the so-called. “machine learning” (Machine Learning). In other words – the AI engine has analyzed countless resources once created by man and on their basis is able to generate its own work “inspired” by the data it uses.

Four AI-generated images for the phrase “industrial landscape” and four art styles:

For some time, work has been underway to create tools where 3D objects will be created on the same principle. Among the companies developing “text to 3D” systems are e.g. OpenAI (developers of ChatGPT), NVIDIA and Luma AI. While these systems are more of a concept form for now, they do require source material for machine learning. At the end of January this year, we already described the effects of this type of work created by NVIDIA:

Nvidia specialists generated a 3D model of the frog after entering the phrase: “blue frog with poison dart sitting on a water lily”. The obtained object, after applying the necessary improvements, can be used in CGI graphic scenes or video games. Nvidia treats Magic3D as a response to DreamFusion – a system for converting text into 3D objects presented by Google researchers in September 2022.

Nvidia’s research and development team argues that their technology will allow anyone to create 3D models without the need for special training. “When refined, the resulting technology could accelerate the development of video (and VR) games and perhaps eventually find application in special effects for movies and television. We hope that with Magic3D we can democratize 3D synthesis and open up everyone’s creativity in creating 3D content.”

Magic3D uses a two-step method that takes a rough model created at low resolution and optimizes it to a higher resolution, using a text-to-image model to produce a 2D image. This is then optimized to a volumetric NeRF (Neural Radiation Field). According to Nvidia, the Magic3D modeling technology allows you to generate them twice as fast as DreamFusion.

Magic3D can also perform quick 3D mesh editing. Given a basic prompt and a low resolution 3D model, the text can be modified to alter the resulting model. Additionally, the authors of Magic3D illustrated keeping the same output design through multiple iterations and implementing a 2D image style (such as Cubist painting) into a 3D model.

Sketchfab tries to prevent this – or at least influence how its resources (or user resources) will be used by creators of AI tools. By introducing new provisions to the regulations, the company informs:

If you want to protect your files from being used by generative AI programs, you can mark your models as “NoAI”.

You (and other users) agree not to use uploaded files marked “NoAI” in datasets for development or input into generative artificial intelligence programs.

If you share your work under a license that allows redistribution (including as part of a larger work), such as one of the Creative Commons licenses or the Sketchfab Standard License Agreement, you agree to mark all of your models that have been made using generative AI programs as “CreatedWithAI”.

Sketchfab agrees not to use any data you submit in datasets to develop AI generative programs or as input to those programs.

Sketchfab agrees not to license any files you submit to third parties for use in datasets, development, or as input to generative AI programs.

Source: www.sketchfab.com

Although, given that the results achieved by AI tools for generating 3D objects based on a text description are rather in their infancy at the moment, Sketchfab’s step towards making users aware of the risk of sharing their resources for free, e.g. for the purpose of machine learning seems to be justified. If similar steps were once taken by the largest photo and image repositories, many graphic designers and photographers would not have to worry about their jobs or future career today…

Source: www.sketchfab.com via www.fabbaloo.com
Image: 3D printers generated by one of the leading AI tools available on the internet