← BACK_TO_LOG

OpenAI Adds SynthID and a Public Verification Tool

2026-05-22 · openai

On May 19, 2026, OpenAI expanded its provenance stack with C2PA conformance, SynthID watermarking for images generated through ChatGPT, Codex, or the API, and a preview of a public verification tool. The release is meant to make AI-generated media easier to identify and verify across tools.

Key Features or Updates

OpenAI is combining signed metadata with invisible watermarking so provenance can survive more real-world transformations. The new verification preview is meant to help people check whether an image came from OpenAI tools and whether provenance signals are still present.

Impact on Developers

For teams shipping media features, provenance is becoming part of the product contract. Apps that ingest, transform, or redistribute AI-generated images should preserve these signals instead of flattening them away.

How to use it

If your stack handles AI images, verify how uploads, conversions, and exports affect Content Credentials. Start surfacing provenance status in your UI and storage layer now so you can adapt as other platforms adopt the same signals.

Read Original Post →