OpenAI has launched a free tool that helps users check whether an image was created with its AI systems. Available at OpenAI Verify, the tool detects two markers: C2PA metadata and SynthID watermarks in images from ChatGPT, the OpenAI API, and Codex.
C2PA metadata records details like the image’s origin, creator tool, and timestamp, but it can be removed when files are edited or shared online. To make detection more reliable, OpenAI also uses Google DeepMind’s SynthID, an invisible watermark designed to survive screenshots, resizing, and compression.
OpenAI says the two systems work together: C2PA provides detailed provenance information, while SynthID offers stronger durability against tampering.
The move comes ahead of new AI transparency laws taking effect in the EU and California on August 2, 2026. Companies including Kakao, ElevenLabs, and Nvidia also announced support for C2PA and SynthID, while Google plans to expand detection across Search and Chrome.
The tool currently only identifies content generated by OpenAI systems and cannot verify images from platforms like Midjourney or Stable Diffusion. OpenAI describes it as a public preview with no confirmed full-release timeline.
