Adobe, Nvidia AI imaging systems aim to resolve copyright questions | – #Adobe #Nvidia #imaging #systems #aim #resolve #copyright #questions

March 21 (Reuters) – Two Silicon Valley companies on Tuesday announced new tools to create images using artificial intelligence as they tackle the most vexing legal issues surrounding the technology: copyright and fees.

Adobe Inc ( ADBE.O ) has added artificial intelligence to its most popular software, including Adobe Photoshop and Adobe Illustrator, to speed up the process of creating images and text effects, noting that creators whose work has been used by the tools can get paid.

Nvidia Corp ( NVDA.O ) has unveiled its own service, known as Picasso, which uses artificial intelligence to create images, videos and 3D applications from textual images. Nvidia trained the technology on licensed images from Getty Images, Shutterstock Inc ( SSTK.N ) and Adobe and plans to pay royalties.

It is a milestone in the ongoing tension between the rights of copyright holders and emerging technology. Image generation technology is “trained” on billions of images, but it’s not always clear whether this use is legally permitted.

Earlier this year, Getty Images sued Stability AI, the creators of open-source art generation software, alleging that Stable Diffusion copied more than 12 million images from its database without permission.

“This collaboration (with Nvidia) is a testament to the feasibility of a responsible AI development path and the unique nature of Getty Images content and data,” Getty Images CEO Craig Peters told Reuters in an email.

“This is consistent with our belief that generative AI is an exciting tool that should be based on permissioned data, visuals and individual privacy.”

Adobe’s new AI-powered feature called Firefly lets users use words to describe images, illustrations or videos the software will create. According to him, the AI ​​was trained on Adobe Stock images, open-licensed content and older content whose copyrights have expired, resulting in works that are safe for commercial use.

The company also advocates for a universal “do not train” label, which allows photographers to request that their content not be used to train models.

“We are very interested in making this creator a friend,” Ely Greenfield, chief technology officer for digital media at Adobe, told Reuters.

If Adobe users ask the system for an image in the style of a particular artist, “it won’t produce an image that matches that person’s style,” Greenfield said. “You can sell that as an artist. If someone wants to use your style, you can actually sell the customer the right to use your style.”

Nvidia’s Picasso AI-image generator is part of a collection of AI-powered cloud products unveiled at the GTC Developer Conference.

“It’s key to having something that’s going to be interesting to the market,” said Greg Estes, Nvidia’s vice president of developer programs for working with partners like Getty.

“Because other software providers or any enterprise don’t want to get involved (with image-generating AI) without knowing what the origin of the underlying training images is,” he said.

Jun-Yan Zhu, an associate professor at the Carnegie Mellon Robotics Institute, said it’s not uncommon for open-source AI models to train on billions of images. A number of factors determine whether photographers know their work is being sampled, including whether a photographer is famous or whether the training dataset is publicly available.

Zhu said he hopes photographers and artists can ultimately benefit by using the technology to license their artistic styles.

“The livelihoods of content creators depend on respect for intellectual property rights and the value of their creative endeavors,” said Getty’s Peters.

“We believe that innovation and creativity should be fair for the work of artists, photographers, videographers and creatives everywhere. compensation thrives in an environment where it can be used, especially when used for commercial purposes.”

(This story was corrected to say that Adobe’s rendering tools were not developed in collaboration with Nvidia in paragraph 3)

Reporting by Stephen Nellis in San Francisco and Dawn Chmielewski in Los Angeles; Edited by Richard Chang

Our standards: Thomson Reuters Trust Principles.

2023-03-22 07:58:21
Source – reuters

Translation“24 HOURS”