OpenAI has unveiled DALL-E3, the latest version of its text-to-image AI tool. DALL-E 3 comes with a significant improvement over its predecessor, DALL-E2, and can generate highly realistic and detailed images from text descriptions, claims its makers.
DALL-E3 is trained on a massive dataset of text-image pairs, and it can generate images from a wide range of prompts, including object description, scenes, and abstract concepts.
One of the key differences, highlighted by OpenAI, between modern text-to-image systems and DALL- E3 is the need for smart prompts. Unlike modern text-to-image systems that have a tendency to ignore words or descriptions, forcing users to learn prompt engineering, DALL-E3 can generate images that exactly adhere to the text you provide.
Dall- E3 comes with ChatGPT integration that automatically generates tailored, detailed prompts for DALL-E3. For example- if you like a particular image, but it’s not quite right, you can ask ChatGPT to make tweaks with just a few words.
ALSO READ l Meta for India: Mark Zuckerberg announces WhatsApp Flows, Meta Verified launch timeline, other updates
DALL-E3 will be available to ChatGPT Plus and Enterprise customers in early October
OpenAI states that it has taken steps to limit DALL·E 3’s ability to generate violent, adult, or hateful content. Furthermore, the image creator has mitigations to decline requests that ask for a public figure by name. The company has partnered with Red Teamers, domain experts who stress-test the model, to help inform DALL- E3 risk assessment and mitigation efforts in areas like propaganda and misinformation.
OpenAI is also internally testing ways to help people identify when an image was created with AI.
“We’re experimenting with a provenance classifier—a new internal tool that can help us identify whether or not an image was generated by DALL·E 3—and hope to use this tool to better understand the ways generated images might be used,” it informs.
Follow FE Tech Bytes on Twitter, Instagram, LinkedIn, Facebook.
