Stock photo provider Getty Images released its own AI image generator called ‘Generative AI by Getty Images’ that only uses content from the company’s media library. It said people can use the AI to generate and improve images with “full protection and usage rights.”
In a blog post this week, Getty Images said the new tool uses Edify, a custom machine learning model under the generative AI foundry NVIDIA Picasso, and is trained solely from the company’s “creative library, including exclusive premium content.”
Getty Images tackles AI copyright issues
“We’re excited to launch a tool that harnesses the power of generative artificial intelligence to address our customers’ commercial needs while respecting the intellectual property of creators,” said Craig Peters, CEO at Getty Images.
Generative AI by Getty Images is a significant effort by the Seattle-based company to tackle issues of copyright that have dogged the AI industry, not least the stock photo provider itself, since the launch of the vital chatbot ChatGPT in 2022.
In February, Getty sued Stability AI for copying over 12 million of its photos without a license and using them to train its Stable Diffusion AI image generation system. The firm accused Stability of infringing its copyrights and of competing with it unfairly, Reuters reported.
According to the Generative AI by Getty Images webpage, the new image generator is said to be “commercially safe”, meaning that it comes without “intellectual property or name and likeness concerns, no training data concerns.”
Users can download all the photos they create using the AI and even license them. Getty said there’s an uncapped indemnification that comes with each download, as well as “the right to perpetual, worldwide, nonexclusive use in all media.”
In addition, any content generated via the AI tool will not be added to the Getty Images library for others to license.
“We’ve created a service that allows brands and marketers to safely embrace AI and stretch their creative possibilities while compensating creators for inclusion of their visuals in the underlying training sets,” said Peters, the Getty Images CEO.
Blocking ‘problematic content’
More and more people have resorted to using AI tools such as ChatGPT, Midjourney, and a host of others to create content in the form of text, videos, and images. However, the nascent technology has also raised serious ethical, privacy, and copyright concerns.
Earlier this year, the U.S. Copyright Office ruled that AI-generated images were not protected under copyright law. The landmark ruling came after the author of “Zarya of the Dawn” applied for copyright protection for both text and illustrations in her comic book.
On its website, Getty Images said its partnership with NVIDIA trained the custom AI model “from the very best creative visuals from Getty Images to create commercially safe models and outputs.”
“And we’ve taken care to block prompts that could generate and output problematic content,” it added.
People can access the tool on the Getty Images website. They can also choose to integrate the service into their existing workflows and apps via an API. Getty says users “will soon be able to customize the AI with proprietary data to produce images with their unique brand style and language.”