![Brush illustrating the concept of generative AI art on technological background](https://static.seekingalpha.com/cdn/s3/uploads/getty_images/1624663882/image_1624663882.jpg?io=getty-c-w750)
Cinefootage images/iStock via Getty Images
Microsoft (NASDAQ:MSFT)-supported OpenAI unveiled the latest version of its text-to-image AI tool called DALL·E 3.
The company said DALL·E 3 is now in research preview and will be available in October for ChatGPT Plus and Enterprise customers, via the API and in Labs later this fall.
DALL·E 3 is built natively on ChatGPT, allowing one to use ChatGPT and refine the prompts.
DALL·E 3 understands significantly more nuances and details than the company’s previous systems. Modern text-to-image systems tend to ignore words or descriptions, forcing users to learn techniques quickly. But DALL·E 3 is a leap forward in its ability to generate images that precisely match the text a user provides, according to OpenAI.
However, the company noted that, like previous versions, it has taken steps to limit DALL·E 3’s ability to generate violent, adult, or hateful content.
The new version includes measures to reject requests asking for the name of a public figure. The company said it has improved safety performance in risk areas such as public figure generation and harmful biases related to visual over/underrepresentation, working with experts to help inform its mitigation efforts in areas such as propaganda and disinformation.
In addition, OpenAI noted that it is also researching the best ways to help people identify when an image was created with AI. The company is experimenting with a provenance classifier – a new internal tool to help identify whether an image was generated by DALL·E 3 – and hopes to use this tool to better understand the ways in which generated images can be used.
DALL·E 3 is also designed to reject requests asking for an image in the style of a living artist.
OpenAI’s text-to-image AI tools also have competitors such as Alibaba’s (BABA) Tongyi Wanxiang, Midjourney Inc.’s Midjourney and Stability AI.
OpenAI is also facing criticism from authors including George RR Martin and John Grisham, who have sued the company for alleged copyright infringement.
Generative AI services have taken the world by storm since the launch of ChatGPT. Companies around the world have launched their own major language models, or LLMs. Baidu’s (BIDU) Ernie Bot, Alibaba’s (BABA) Tongyi Qianwen and Tongyi Wanxiang, The Alphabet unit (GOOG) (GOOGL) Google’s Bard, Meta Platforms (META) AudioCraft, SeamlessM4T and Llama 2 are some of them.