EXPLORING THE AI WORLD 
High-profile tech company NVIDIA has recently announced the launching of Nvidia AI Foundations, a cloud services platform for customers to train and deploy generative AI models using custom datasets.
And they already have partnerships with important media-focused firms, such as Shutterstock, Adobe, and Getty Images, to train different AI image generators on the new NVIDIA clouds. Also, BRIA, an AI generative model developer, announced all its models are available on the new cloud services.
Interested? All the details are below!
The new cloud services under the brand NVIDIA AI Foundations are actually two: NVIDIA NeMo for language-generating models such as chatbots and customer support assistants and NVIDIA Picasso for models that generate image, video, and 3D content.
Through these services, developers, and enterprises can use their custom data to train and test their own AI-generative tools. And they can also use pre-trained models by NVIDIA. Once trained and tested, the AI apps can be run on these very clouds or anywhere else.
Currently, the NeMo cloud is in early access, and Picasso is in private preview. Both services are available upon application and pending approval.
The tech giant expects this kind of service to contribute to the scalability of AI-generative models and produce legally safer applications trained with approved and curated data only.
The stock photo titan Shutterstock disclosed they are teaming up with NVIDIA AI Foundation on the Picasso cloud to train an upcoming, state-of-the-art generative tool that can transform text prompts into high-fidelity 3D assets.
Its capabilities will include “create object shapes, assist with object unwrapping, generate textures and materials and, for non-3D users, [it will] produce complete 3D models ready for use in a wide variety of applications and platforms,” according to the agency.
The 3D graphic generator will be trained solely with content from Shutterstock’s libraries –making it legally safer– and contributors whose work was included in the training datasets will be compensated via the Shutterstock Contributor Fund.
Once ready, the novel app will be available on the Shutterstock Creative Flow platform –where it will complement Shutterstock Generate, the current AI image generator built by Shutterstock in collaboration with OpenAI– as well as on TurboSquid –Shutterstock’s subsidiary for 3D media assets– and even on the NVIDIA Omniverse™ platform for industrial metaverse applications.
Another big announcement regarding partnerships is that Adobe and NVIDIA are working to co-develop commercially sustainable AI image-generation products on the Picasso cloud based on the newly launched Adobe Firefly family of AI-generative models.
They will collaborate to produce text-to-image, text-to-video, and text-to-3D applications, all framed in Adobe’s Content Authenticity Initiative (CAI) that seeks to establish new, ethical industry standards for AI-generated media.
These new applications will be integrated into flagship Adobe software, such as Photoshop and Premiere Pro, in an effort to include AI media generation as part of creative workflows.
Another large company using NVIDIA AI training clouds is Getty Images, which will use its own licensable content to train proprietary text-to-image and text-to-video generative features.
Since Getty Images is currently suing Stable Diffusion for copyright infringement, alleging unauthorized use of Getty Images’ media files in training its software models, it’s very relevant that they are now building their own legally-backed tools using NVIDIA’s cloud capabilities.
Lastly, we also learned that advanced AI generative model developer BRIA has made its responsible, copyrighted AI image generator products available on NVIDIA Picasso, leveraging the cloud service's optimization capabilities to enable enterprise clients to produce AI-generated images and edit high-quality images via AI image editors faster and with better results than ever before.
BRIA's models are all trained exclusively with licensed data, making them more ethically responsible and safer to use.
It's worth noting here that BRIA is already partnered with Getty Images to bring its AI image editors into the stock photography platform's toolkit. BRIA's ethical approach to AI visual generation was one of the reasons for this teaming up.
These announcements make clear that such cloud services by NVIDIA are already helping scale up the production of AI generative tools for end users that are very promising in terms of functionality and legality.
3D asset generation and video generation, which are yet pretty much uncovered for customers but considerably explored by developers, are already being worked on, on NVIDIA Picasso and from high-profile firms who know all there is to know about copyright, commercial use, and image quality.
It’s undoubtedly exciting!
THE AUTHOR
Ivanna Attie
I am an experienced author with expertise in digital communication, stock media, design, and creative tools. I have closely followed and reported on AI developments in this field since its early days. I have gained valuable industry insight through my work with leading digital media professionals since 2014.
AI Secrets is a platform for tech decision-makers to learn about AI technology. Our team includes experts such as Amos Struck (20+ yrs ICT, Stock Photo, AI), Ivanna Attie (expert in digital comms, design, stock media), and more who share their views on AI.
Get AI news in your inbox & join thousands of engineers and managers using AI to boost sales and grow market share
