OpenAI’s DALL-E to Utilize Shutterstock’s Image Library for Additional Six Years

OpenAI’s DALL-E, a cutting-edge text-to-image model, will continue its training on Shutterstock’s vast library of images for an additional six years. The collaboration between OpenAI and Shutterstock began in 2021 when the stock image site allowed OpenAI to utilize its images for training DALL-E. OpenAI CEO Sam Altman hailed this partnership as crucial for the model’s development. Shutterstock also initiated a Contributor Fund last year to compensate artists when their work is used to train OpenAI’s models. Alongside this, Shutterstock integrated OpenAI’s image generator directly into its website and prohibited the sale of AI-generated images that were not produced using its built-in DALL-E tool. Now, Shutterstock is expanding this integration by granting users the ability to manipulate and transform any image in its extensive library. Furthermore, Shutterstock has plans to introduce AI features to Giphy, the popular GIF-making platform it recently acquired from Meta.