Getty Images Launches AI-Powered Image Generator

Getty Images has unveiled a new product powered by artificial intelligence that enables the generation of images based on textual prompts. According to Bloomberg, these images will be generated from a vast pool of content owned by Getty Images itself.

The new Getty product, developed in collaboration with chip manufacturer Nvidia, will be trained on Getty Images' own content and will seek to navigate complex legal issues, including limiting the types of images the generator can produce. Previously, Getty Images had filed lawsuits against companies that used their images without proper authorization for training AI models and generating images.

Getty Images stated that this tool will exclusively use creative content, not news photographs, to prevent the creation of deepfakes. The new image generator will also prohibit the use of trademarked materials or assets that users do not own. As a result, this image generator won't be capable of creating something akin to a viral image of Pope Francis wearing a Balenciaga coat.

The company also promises to compensate the authors whose content was used in training the AI models.

Getty Images will allow customers to add their own data and brands to the images, making this generator suitable for businesses to advertise their products. However, content generated with this tool will not be included in the Getty Images library, making it inaccessible to other platform users. Images generated by artificial intelligence will have Getty licenses for usage and protection against potential legal actions.

As a reminder, George Martin and a group of other renowned science fiction writers filed a lawsuit against OpenAI. The plaintiffs claim that their copyrighted works were used for AI training without their permission. According to them, the company utilized their works to develop large language models and algorithms that provide text responses to user queries, mimicking human language.  The authors of the lawsuit demand a court injunction against the use of their works in large language models without their permission and seek compensation for damages. This lawsuit is in line with similar cases where authors are protecting their copyright. In July of last year, Sarah Silverman and two other authors filed complaints against OpenAI and Meta for copyright infringement.