AI Watermark Center
OpenAI research

OpenAI Content Credentials

What C2PA Content Credentials can contain on supported OpenAI images, how they differ from SynthID, and which formats this site can inspect locally.

Published
Last verified
Review
Reviewed against primary sources

Content Credentials are C2PA provenance metadata. SynthID is an imperceptible watermark in the media itself. OpenAI documents both on supported current images. Finding C2PA is not a SynthID detection, and this inspector is not an OpenAI SynthID detector.

Current OpenAI image scope

Help Center documentation currently lists C2PA metadata on supported images generated with ChatGPT, Codex, and the OpenAI API. OpenAI also historically added C2PA to DALL·E 3 images in 2024. Historical launch coverage is not a licence to say every file since 2024 still carries a readable manifest.

What issuer, model, and generation fields may contain

OpenAI’s Content Provenance API documentation describes image C2PA results that can include issuer, model, generation time, and a validation state when those fields are present. Presence of a manifest is not proof of accuracy, legal ownership, or an unedited file. Identification of OpenAI as issuer depends on the credential data, not on “C2PA exists.”

Metadata durability

Screenshots, some social uploads, transcodes, and editor re-exports can drop C2PA. Missing C2PA does not mean the file is not OpenAI-generated. Absence of embedded credentials is not absence of SynthID. This page does not teach credential stripping for concealment.

What AI Watermark Center can inspect

Embedded manifests in JPEG, PNG, and WebP in the browser. Soft bindings and cloud recovery are not performed. C2PA inspection does not detect SynthID. Official OpenAI verification remains a separate, external provider tool.

Inspect Content Credentials

Official OpenAI verification (opens in a new tab)External provider tool

Sources

  1. Provenance signals (Content Credentials, SynthID) in OpenAI-generated content — OpenAI Help Center

    Accessed August 17, 2026.

    Primary source for current OpenAI image C2PA + SynthID on supported ChatGPT, Codex, and API outputs; audio SynthID; prompt-requested visible disclosure; and openai.com/verify. Visible marks are separate from embedded signals.

  2. Content provenance — OpenAI API documentation

    Accessed August 17, 2026.

    Content Provenance API (also referenced as OPENAI-API-PROV-001 in Phase 7 notes). Images: C2PA + SynthID. Audio: SynthID. Documents not_detected semantics. Public/documented; not integrated by AI Watermark Center.

  3. Understanding the source of what we see and hear online — OpenAI

    Accessed August 17, 2026.

    Historical: OpenAI began adding C2PA metadata to DALL·E 3 images in ChatGPT and the API in 2024. Chronology only — not current product-scope documentation. Also listed as OPENAI-IMAGE-HISTORY-001 in Phase 7 source notes.

  4. Verify — OpenAI

    Accessed August 17, 2026.

    Public verification surface referenced by OpenAI Help Center for supported image and audio provenance signals (C2PA and/or SynthID). External provider tool — not an AI Watermark Center feature.

In this cluster