OpenAI uses generative AI image discovery tool

lPeople are adopting generative artificial intelligence to create and edit images, video and audio in ways that enhance creativity, productivity and learning, with the audio-visual content created becoming increasingly ubiquitous.“says the company, explaining that the advent of DALL-E3 software (a generative artificial intelligence program capable of creating images from text descriptions) has made differentiating synthetic images increasingly difficult.

The startup behind acclaimed generative AI tools ChatGPT and DALLE-E announced in a blog post the launch of a tool that can detect whether an image was created using the SLAB-E3 image generator.

“We are introducing new tools to help researchers investigate the authenticity of content.”writes a launch after running internal tests on an earlier version of the tool.

According to the company, the tool correctly identified images generated by DALL-E 3 approximately 98% of the time in internal testing and could handle common changes such as compression, cropping, and saturation changes with minimal impact.

Startup OpenAI is following in the footsteps of tech giants Meta and Google by joining C2PA, a coalition to verify the origin and authenticity of content.

(with CARD)