Shutterstock rolls out AI image generator tool powered by DALL-E

Shutterstock, a leading provider of stock photos, 3d models, and templates, has released an Artificial Intelligence-based (AI) image generator tool that can convert text instructions into images. Shutterstock announced last October that it will integrate OpenAI's DALL-E 2 with its own content. 

Shutterstock said on Wednesday that it has been working for the last two years with OpenAI, Meta, and LG AI Research to build its own generative AI capability. OpenAI’s CEO Sam Altman confirmed last October that DALL-E 2 was trained using Shutterstock data and it played a critical role in its development. 

Shutterstock claims its AI image generator can generate unique images from a single-word instruction. It is available on Shutterstock website under the ‘Images’ section. 

AI-based image generation models such as OpenAI’s DALL-E and StabilityAI’s Stable Diffusion have become very popular with Internet users. Several apps such as Prisma Labs Lensa and Deviant Art have integrated these models to offer image-generation tools to their users. These tools can generate complex artwork and images from text or selfies within seconds. 

"Our easy-to-use generative platform will transform the way people tell their stories. You no longer have to be a design expert or have access to a creative team to create exceptional work,” Paul Hennessy, CEO of Shutterstock said in a statement. 

Hennessy added that it is built on an ethical AI approach and on a library of assets that represents a diverse world.  

Bias in AI models is a top concern for researchers and companies building AI solutions. According to a Wired report, published last May, OpenAI in its tests found that DALL-E 2 was enforcing racial and gender stereotypes by generating images of white men by default and overly sexualizing images of women.

Early this month, a group of artists and lawyers filed a lawsuit against Stability AI, Deviant Art, and Midjourney for using copyright images of artists without consent to train sta­ble dif­fu­sion. Their lawsuit claims that the image generative model will flood the mar­ket with unlim­ited num­ber of copied images and will cause significant damage to artists and the art market.