Getty Introduces AI Image Generator on iStock with Comprehensive Legal Protections for Users

Getty Images, a leading provider of stock imagery, today advanced its AI initiatives by launching a specialized image generation service for its subsidiary, iStock.

Named Generative AI by iStock, this innovative offering equips designers and business users with a text-to-image generator, enabling them to create ready-to-use visuals from simple text prompts.

A standout feature of this tool is that the generated images come with legal protections and usage rights, a significant advantage over many existing image generators. This means businesses can safely utilize these images for commercial purposes, such as marketing campaigns, without the risk of unauthorized use.

Today's announcement highlights Getty's second major step in the growing generative AI landscape, following the launch of an image generator for its primary platform in September. This development follows Shutterstock's introduction of a similar AI image generation feature in fall 2023.

What to Expect from Generative AI by iStock

Debuted at CES 2024 in Las Vegas, Generative AI by iStock mirrors the image generation tool introduced by Getty. Users will be able to describe their desired visuals, and the advanced model will produce photo-quality images in resolutions up to 4K.

Leveraging Getty's extensive catalog of licensed images, this tool is trained on proprietary creative content, including exclusive photos, illustrations, and videos. This training ensures that users receive safe, commercially viable AI-generated visuals, especially in light of ongoing legal battles concerning AI tools’ usage of artists' works without consent. Getty has notably filed lawsuits against Stability AI over similar issues.

While it remains uncertain if training on stock content impacts output quality, consumers can take comfort in the fact that each image generation will produce four distinct visuals, each backed by legal coverage of up to $10,000. This indemnification protects users against potential content theft, and iStock assures that the generated content will not be added to its library for external licensing.

The tool features a user-friendly prompt builder, allowing users to specify various inputs, including the subject, its actions, surroundings, color, mood, and aspect ratio.

Powered by Nvidia Picasso Model

The entire image generation process utilizes Nvidia’s Picasso AI model library. Getty has integrated Picasso’s Edify foundation model for visual design with its licensed content, resulting in a custom model that drives Generative AI by iStock.

Accessible starting today, the service features a structured pricing plan for 100 image generations. Additionally, Getty is rolling out advanced editing capabilities, including inpainting and outpainting. Inpainting enables users to mask specific areas of an image and seamlessly replace them with objects specified through text. Outpainting allows users to expand images to fit various aspect ratios, filling in content based on the original.

In the competitive AI-driven content generation market, Getty’s tools position themselves against key players like Midjourney, DeviantArt, Runway, and Stability AI, while Adobe has introduced its own Firefly model, trained on its exclusive catalog of licensed images, within several Creative Cloud applications.