AI apps ChatGPT images
What ChatGPT writes into your images
Images made with ChatGPT and OpenAI’s image models carry C2PA Content Credentials that say they were AI-generated. Check a file and see what platforms will show.
Read in this browser tab. Never uploaded.
The short version
- Content Credentials
- OpenAI adds C2PA Content Credentials to images from ChatGPT and its API, declaring them as AI-generated.
- What platforms do
- Platforms that read C2PA, such as Instagram, TikTok and LinkedIn, can label these images automatically.
- When it disappears
- Screenshots and many messaging apps strip the credentials, so a copy may show nothing.
Based on public announcements, last reviewed November 2026. Rules change; if this is out of date, tell us on GitHub.
Tools that help
Remove private data
Rewrite your own photos without the details you don’t want to share. No quality loss.
- Photo metadata removerStrip location, camera, editing history and comments in one go.
- Remove location from photoDelete the GPS position but keep camera and date details.
- EXIF removerRemove the EXIF block only: camera, serials, dates and GPS.
- XMP removerRemove editing history, app names and document IDs.
- Hide your AI promptRemove the prompt and settings Stable Diffusion or ComfyUI saved in a PNG.
Check a file
Read what a photo or video says about itself before you post or trust it.
- What’s in my photo?See every piece of hidden data a file carries.
- AI image checkerFind any AI declaration: Content Credentials, IPTC field or prompt.
- Content Credentials checkerRead and verify C2PA credentials locally: tool, edits, signer.
- Photo privacy checkRate what a photo gives away: location, serials, owner name.
Understand labels
What file data can and can’t tell you about AI.
How platforms label AI
What AI apps write into files
ChatGPT images, Adobe Firefly and Photoshop, Google Gemini and Imagen, Meta AI, Stable Diffusion and ComfyUI