Getty Images switches sides: from suing to licensing with generative AI

For most of the generative AI era, Getty Images has been the most determined opponent in court, suing the creators of image generators rather than licensing its content to them. However, on Monday it switched sides, and the market rewarded this shift with a significant increase in its stock value, unusual for a stock photography company. This move not only reflects a pragmatic business strategy but also marks a milestone in the evolution of the relationship between copyright holders and artificial intelligence companies.

What happened?

Getty Images has announced a multi-year licensing agreement with OpenAI, the company behind ChatGPT and DALL-E. This agreement allows OpenAI to use Getty's vast image catalog — which includes over 477 million assets, according to company data — to train its generative AI models, including DALL-E 3 and potential future versions. In return, Getty will receive financial compensation, though exact terms have not been disclosed. According to The Next Web, the news caused Getty Images shares to rise up to 20% on the New York Stock Exchange, closing the session up 18.5%, representing a market capitalization increase of approximately $150 million. This surge is particularly notable for a stock photography company, whose shares are usually stable and low volatility.

For context, Getty Images had maintained a combative stance since 2022, when it filed a lawsuit against Stability AI for unauthorized use of its images in training Stable Diffusion. The lawsuit, still ongoing, alleges infringement of over 12 million photographs and trademark violations. In contrast, the agreement with OpenAI represents a 180-degree turn: from legal confrontation to commercial collaboration.

Why is this important?

This agreement marks a turning point in the relationship between visual content companies and generative AI companies. Until now, Getty had adopted an aggressive posture, suing Stability AI and also other players like Midjourney, though that lawsuit was partially dismissed. The shift toward collaboration suggests that content companies are seeking new revenue streams in the AI era, rather than simply opposing it. According to a Bloomberg Intelligence analysis, the market for data licensing for AI could reach $1.5 billion by 2027, incentivizing companies like Getty to participate.

Moreover, the agreement has strategic implications for OpenAI. The company has faced criticism and lawsuits from authors, artists, and media outlets for using copyrighted content without permission. By obtaining a license from Getty, OpenAI legitimizes its approach and reduces the risk of future litigation. This move adds to similar agreements with Shutterstock (announced in July 2023) and The New York Times (ongoing), showing a trend toward collaboration rather than confrontation.

Consequences for the industry

• Legal and commercial precedent: This agreement could set a precedent for other stock companies, like Shutterstock, which has already signed similar deals with OpenAI and Meta. It could also influence future court decisions on the use of copyrighted content for AI training, such as the ongoing case of The New York Times against OpenAI. Legal experts note that agreements like this demonstrate that the market can self-regulate, which could lead courts to favor commercial solutions over infringement penalties.
• Impact on creators: Photographers and artists who contribute to Getty may see changes in how their work is compensated when used for AI training. Getty has announced it will create a fund to compensate contributors, though details are scarce. In comparison, Shutterstock has already implemented a royalty system for its contributors for the use of their images in AI training. Getty is likely to adjust its revenue-sharing policies to maintain the loyalty of its creator community, which in some cases has expressed concern about lack of transparency.
• OpenAI's strategy: For OpenAI, this agreement legitimizes its approach of obtaining licenses for training content, which could help mitigate future lawsuits and criticism over the use of protected material. Additionally, it provides access to a high-quality, curated catalog, improving the quality of its models. In the past, OpenAI has faced criticism for using data obtained in questionable ways, such as in the case of J.D. Vance's book 'The Line,' which was used without permission. This agreement reinforces its commitment to ethical practices, at least in the visual domain.

What readers should know

This move does not mean Getty Images has completely abandoned its legal stance. The company continues its lawsuit against Stability AI, which is expected to go to trial in 2025. However, the agreement with OpenAI shows it is willing to collaborate when conditions are favorable. For generative AI users, this could translate into more robust models and fewer legal issues in the future. Additionally, we are likely to see more agreements of this kind as the industry seeks a balance between innovation and copyright.

In market terms, the positive investor reaction suggests they see the agreement as a validation of Getty's business model in the AI era. The company's shares have risen over 40% year-to-date, far outperforming the S&P 500. However, analysts warn that the real revenue impact will depend on the scale of the agreement and Getty's ability to monetize its catalog on a recurring basis.

In conclusion, the Getty-OpenAI agreement represents a seismic shift in the stock photography and generative AI industry. It demonstrates that even the most affected industries can find ways to coexist and thrive, provided they are willing to adapt. As Craig Peters, CEO of Getty Images, said in an interview with The Verge: 'It's not about fighting AI, but about finding a way to work with it that benefits everyone, including creators.'

"Getty Images has gone from being the fiercest adversary of generative AI to a business partner, proving that even the most affected industries can find ways to coexist and thrive."