ChatGPT is beginning to create animated GIFs as part of OpenAI’s Images 2.6 update, announced on November 4. Users can request a looping animation or start with a still image, extending the image tool from static artwork to short sequences of motion.
In its launch announcement, OpenAI describes uses including animated characters, product loops and storyboards. The company says Images 2.6 generates images 50% faster than its predecessor and preserves visual details more consistently during edits. Those are vendor claims, rather than results from independent DaySec testing.
What users can do
The official image documentation says users can describe what should move or upload a still image and ask ChatGPT to animate it. Changes to a generated GIF can be requested in the conversation. Opening the result provides controls to change playback speed, save it or share it.
There is a practical limit: ChatGPT cannot edit an animated GIF uploaded by the user. Animating a still picture and revising an animation created inside ChatGPT are supported workflows; importing an existing animated file for editing is a separate capability.
OpenAI says Images 2.6 is available globally across subscription tiers in ChatGPT and Codex. Its ChatGPT release notes nevertheless describe GIF creation as rolling out. Availability of the image model should therefore not be read as confirmation that every account already has the GIF feature.
Motion changes the moderation challenge
For a security and trust audience, the relevant change goes beyond convenience. OpenAI’s Images 2.6 system card says harmful meaning can emerge across animation frames even when an individual frame does not reveal it. The company has added GIF-specific safeguards, including broader restrictions on some content, but acknowledges that motion-related violations remain difficult to detect.
The same document describes provenance measures using C2PA metadata and invisible SynthID watermarking across ChatGPT, Codex and the API. OpenAI presents these as complementary tools and says provenance has no single complete solution. The launch expands creative options while leaving an ongoing task for moderation systems: judging the sequence as well as the picture.
