OpenAI is preparing to launch DALL-E 3, the third version of its text-to-image AI system. The new version improves upon its predecessor’s capabilities, can generate results within the ChatGPT app, and supposedly has a better understanding of user intentions, especially with detailed prompts. In a demonstration to The Verge, the AI generated four images from a detailed prompt within ChatGPT. Furthermore, DALL-E 3 is designed to excel at creating elements that previous versions struggled with, such as hands and texts in images. It also features enhanced security measures, such as ignoring explicit or hateful prompts. OpenAI has also taken steps to avoid potential legal issues by enabling the AI to decline requests to mimic living artists’ work and allowing artists to request the removal of their work. DALL-E 3 will be released to ChatGPT Plus and enterprise customers next month, and will be available to OpenAI’s research labs and API customers later this year. The company has not yet announced when it will be available to the general public.