DALL·E 3 system card

OpenAI Blog Models
image-generation text-to-image ai-safety red-teaming system-card

Summary

OpenAI released the system card for DALL·E 3, detailing safety evaluations, red teaming efforts, and mitigations implemented before deployment of the improved text-to-image model.

No content available
Original Article Export to Word Export to PDF
View Cached Full Text

Cached at: 04/20/26, 02:51 PM

# DALL·E 3 system card Source: [https://openai.com/index/dall-e-3-system-card/](https://openai.com/index/dall-e-3-system-card/) OpenAI## Abstract DALL·E 3 is an artificial intelligence system that takes a text prompt as an input and generates a new image as an output\. DALL·E 3 builds on DALL·E 2 by improving caption fidelity and image quality\. In this system card, we share the work done to prepare DALL·E 3 for deployment, including our work on external expert red teaming, evaluations of key risks, and mitigations to reduce the risks posed by the model and reduce unwanted behaviors\. - [DALL·E](https://openai.com/research/index/?tags=dall-e) - [Language](https://openai.com/research/index/?tags=language) - [Ethics & Safety](https://openai.com/research/index/?tags=ethics-safety) - [Learning Paradigms](https://openai.com/research/index/?tags=learning-paradigms) - [System Cards](https://openai.com/research/index/?tags=system-cards)

Similar Articles

DALL·E 3 is now available in ChatGPT Plus and Enterprise

OpenAI Blog

OpenAI announces DALL·E 3 is now available in ChatGPT Plus and Enterprise, featuring improved image quality, better text/hand/face rendering, and enhanced prompt adherence through training on improved captions. The release includes multi-tiered safety systems, artist style protections, and an internal provenance classifier for detecting AI-generated images with >99% accuracy.

DALL·E 2 research preview update

OpenAI Blog

OpenAI announces an expansion of DALL·E 2 research preview access, sharing safety metrics and learnings from 3 million images created by early users. The company plans to onboard up to 1,000 new users weekly while continuing to refine content policy enforcement and address training data biases.

OpenAI o3-mini System Card

OpenAI Blog

OpenAI releases the o3-mini System Card, documenting safety evaluations and risk assessments for their advanced reasoning model trained with reinforcement learning. The model achieves state-of-the-art safety performance on certain benchmarks and is classified as Medium risk overall under OpenAI's Preparedness Framework.

DALL·E API now available in public beta

OpenAI Blog

OpenAI announces DALL·E API is now available in public beta, allowing developers to integrate image generation capabilities directly into their applications. Early adopters include Microsoft, CALA, and Mixtiles, with built-in safety features and content moderation.

DALL·E: Creating images from text

OpenAI Blog

OpenAI introduces DALL·E, a 12-billion parameter transformer model that generates images from text descriptions by treating text and images as a single token stream. The model demonstrates diverse capabilities including creating anthropomorphized objects, combining disparate concepts, rendering text, and performing image inpainting tasks.