OpenAI image checks
ChatGPT Image Detector for OpenAI and GPT Images
Checking a picture that may have come from ChatGPT, DALL-E, Codex, or the OpenAI API? This is a ChatGPT SynthID checker: Slop or Not reads OpenAI's SynthID watermark and C2PA Content Credentials first, then scores the pixels with a classifier trained on millions of human and AI images when the proof was stripped. Online checks stay away from outside AI companies.
The only AI image detector that reads Google DeepMind's SynthID watermark from both Google and OpenAI images.
Google's own detector checks only Google's images. OpenAI's checks only OpenAI's. Slop or Not reads both.
What it checks
Check OpenAI Provenance First, Then the Pixels
OpenAI C2PA Content Credentials when they survived export
OpenAI SynthID watermark signals when the image still exposes them
Visual generator artifacts after social apps strip metadata
Good Uses
- Checking a ChatGPT image from a download folder
- Verifying a blog or newsletter image before publishing
- Testing a file after it has moved through chat or social apps
Limits to Know
- OpenAI Verify is scoped to OpenAI provenance signals, not Gemini or every other generator
- Screenshots usually remove C2PA/IPTC metadata and can reduce the strongest provenance evidence
- A clean metadata read does not prove the pixels are human-made
FAQ
ChatGPT Image Detector Questions
Does ChatGPT add metadata to generated images?
OpenAI says images generated with ChatGPT, Codex, and the OpenAI API include both C2PA metadata and SynthID watermarks. Slop or Not checks those signals when the image still exposes them.
What happens if the metadata is gone?
Slop or Not checks for remaining watermark evidence, then falls back to visual image detection and reports a probability. A missing credential is weaker evidence, not proof that the image is real.
When should I use the broader AI image detector?
Use the ChatGPT image detector when OpenAI is the suspected source. Use the broader AI image detector when the generator is unknown or the image has passed through several apps.
Can ChatGPT detect AI images?
Not reliably. ChatGPT can describe an image, but it does not run provenance checks, and its guesses about AI origin are inconsistent. Slop or Not reads OpenAI C2PA Content Credentials and SynthID watermark signals first, then checks the pixels.
Can ChatGPT see or look at images?
Yes. You can attach images and ChatGPT will describe or analyze them. Seeing an image is not the same as verifying it: for a repeatable AI-or-not verdict, run the file through this detector, which checks provenance metadata and watermarks before judging the pixels.
Is this a ChatGPT SynthID checker?
Yes. OpenAI writes a SynthID watermark into the images ChatGPT generates, and Slop or Not reads it, alongside OpenAI's C2PA Content Credentials. It reads SynthID from Google's image models too, so the same check covers both providers. A detected mark can tell you which provider's watermark is present, never what the mark says.
Run the Next Check on Your iPhone or Mac
The Slop or Not app checks images and text on-device, so nothing you check ever leaves your device.
Related checks
Compare Related Detector Workflows
SynthID Detector
Free SynthID detector: check an image for Google and OpenAI SynthID watermark signals online, or run the check on-device on iPhone and Mac.
C2PA Detector
Use this C2PA detector to check Content Credentials. Slop or Not reads signed provenance, then checks pixels when no provenance or watermark resolves the file.
AI Watermark Detector
Use this AI watermark detector and checker to read image SynthID, verified C2PA Content Credentials, and IPTC metadata before any pixel check.