Made with ChatGPT Images 2.0

OpenAI has unveiled ChatGPT Images 2.0, a transformative update to its image generation model that now extends into video creation. The new system, integrated directly into ChatGPT, can produce short, high-quality video clips from simple text prompts, using advanced diffusion architectures for photorealistic rendering. Early demos show smooth animations, consistent character appearances, and detailed backgrounds, with support for resolutions up to 4K. The model handles complex instructions like "a cat jumping off a table in slow motion" with impressive temporal coherence, reducing flickering and artifacts common in earlier AI video tools.

This release is available to all ChatGPT users, including the free tier, though Plus and Pro subscribers get priority access and higher generation limits. OpenAI has also introduced a new interface for iterative editing, allowing users to refine scenes frame by frame. The move positions ChatGPT as a direct competitor to dedicated AI video generators like Runway and Pika, while leveraging its conversational interface for easier control. The update is currently rolling out globally, with OpenAI promising further improvements in video length and resolution in coming weeks.