DALL-E 3 was first introduced on Microsoft’s platform, Bing Chat and Bing Image Generator, even before ChatGPT. However, the technology has faced controversy in the past for outputting copyright image materials, nonconsensual nudes, shifted ethnicity of subjects, and photo-realistic misrepresentations of public figures. OpenAI has taken extensive steps to address these issues and has developed an internal tool capable of 99 percent accuracy in detecting if an image was generated by DALL-E 3.
Key takeaways:
- OpenAI is launching wider availability of its latest text-to-image generator, DALL-E 3, to ChatGPT Plus and Enterprise customers.
- DALL-E 3 improves upon the previous model by allowing users to write longer and more visually descriptive prompts for the image generator.
- Previous text-to-image generators have faced controversy for generating harmful imagery and copyright issues, among other problems.
- OpenAI has taken extensive steps to mitigate these issues with DALL-E 3, including limiting the model's likelihood of generating certain types of content and implementing a tool that can detect if an image was generated by DALL-E 3 with 99 percent accuracy.