TL;DR
- Display Deal: OpenAI and Getty Images have signed a multi-year display agreement for licensed images in ChatGPT search answers.
- Training Caveat: The companies have not disclosed model-training rights, financial terms, or separate contributor economics.
- Getty Scale: Getty works with nearly 600,000 creators and imagery from more than 160,000 annual events.
- User Test: ChatGPT users will need visible image credits, source links, and any later training-rights disclosure.
Getty Images and OpenAI have signed a multi-year display agreement on June 21, giving ChatGPT a licensed image channel for search while leaving model-training rights and deal economics unresolved.
OpenAI can use the deal to show professional photos and editorial images inside ChatGPT search answers if the product rollout follows the agreement. Getty Images, the photo and editorial-media licensing company, has not disclosed financial terms or whether the arrangement includes using images to teach or tune AI systems.
Getty works with nearly 600,000 content creators, almost 360 content partners, and imagery from more than 160,000 news, sport, and entertainment events each year. Craig Peters, Getty Images CEO, framed the user benefit directly: “High-quality, licensed visual content makes AI-powered search and discovery more useful and more trustworthy.”
What the Deal Adds to ChatGPT Search
Agreement terms let ChatGPT show Getty imagery to users while keeping the public scope focused on display rather than model training. Display rights govern what a user sees in an answer; training rights would govern whether image libraries help build or tune the underlying AI system.
ChatGPT Search gives OpenAI the product surface for that change because live answers can carry licensed images instead of pushing users into a separate stock-photo workflow. Live web results already include source links and can surface media directly inside the chat interface.
Getty’s library then becomes part of the visual supply chain for search responses. Users will need visible credits and source links to understand what they are seeing, especially if an answer mixes text, images, and citations in one interface.
Getty shares closed at $1.35 on Monday, up 123 percent from Friday’s $0.61 close after the OpenAI deal became public. Investors treated a display-license announcement as a test of whether Getty can convert its library into AI-search revenue.
Before the jump, Getty shares had been down roughly 55% for the year. Commercial terms remain private because neither company has disclosed pricing, contributor economics, or any separate training-rights payment.
Market Reaction and AI Search Context
In October 2025, Getty gave AI search company Perplexity access to its library through an image-licensing deal. Licensed images can enter search results under that mechanism, and the service has to handle presentation rules around credit and source links.
Perplexity also committed to image credits with direct links to sources, making attribution part of the search result design rather than a separate licensing footnote. OpenAI’s implementation will be judged not only by whether Getty images appear, but also by how ChatGPT identifies where they came from.
Getty has not been simply anti-AI. In September 2023, the company launched Generative AI by Getty Images, and earlier NVIDIA-backed work used fully licensed visual content for text-to-image and text-to-video tools.
OpenAI’s deal extends that controlled-licensing posture into a consumer AI search surface rather than marking a broad surrender of Getty’s rights position. Licensed photos add a rights-cleared visual layer to a search product that already competes on answer quality and source presentation.
Rights Questions Still Define the Deal
Getty’s legal history keeps the training question prominent. In January 2023, Getty sued image-generation company Stability AI over alleged copyright infringement tied to model training.
A later UK proceeding narrowed the case when Getty dropped its primary copyright infringement claims against Stability AI while leaving secondary-infringement and trademark questions alive. Model-training rights remain central to interpreting a display-only OpenAI arrangement.
Getty also continues to bar uploads of user-generated AI images to its own library. Its marketplace policy separates licensed media appearing in AI search from synthetic images entering Getty’s marketplace, even as OpenAI and Perplexity have faced lawsuits over alleged scraping of copyrighted materials.
Implementation will decide whether the arrangement becomes a visible ChatGPT upgrade or a licensing framework that users barely notice. Attribution placement, source links, creator compensation, opt-outs, image-modification limits, and any later training-rights disclosure are the practical checkpoints.
Getty-powered ChatGPT images result will put that implementation in front of users. OpenAI must show visible credit and a source link while any separate training-rights disclosure remains unresolved.