T
traeai
Sign in
返回首页
OpenAI(@OpenAI)

We’re adding new ways for people to identify AI-generated images and understand where they came from...

5.5Score
We’re adding new ways for people to identify AI-generated images and understand where they came from...

TL;DR · AI Summary

OpenAI has added SynthID watermarking to AI-generated images alongside C2PA, but provides no technical details or public access to the verification tool, limiting engineering utility.

Key Takeaways

  • OpenAI embeds SynthID watermark in AI-generated images as a supplement to C2PA C
  • A public verification tool is mentioned to detect OpenAI-generated images, but n
  • No details on watermark robustness, removal resistance, or cross-platform compat

Outline

Jump quickly between sections.

  1. OpenAI adds SynthID watermarking to AI-generated images alongside C2PA Content Credentials to improve traceability.

  2. A public tool is referenced to verify image origin, but no URL, API, or implementation details are published.

  3. No details on SynthID’s algorithm, embedding strength, or tamper resistance are provided for technical evaluation.

Mindmap

See how the topics connect at a glance.

查看大纲文本(无障碍 / 无 JS 友好)
  • OpenAI图像溯源更新
    • 技术手段
      • SynthID水印
      • C2PA Content Credentials
    • 验证机制
      • 公开验证工具(未开放)

Highlights

Key sentences worth saving and sharing.

  • Images now also contain a SynthID watermark, and can be identified using a public verification tool to check whether an image was made by OpenAI products.

    正文第一段

    ⬇︎ 下载 PNG𝕏 分享到 X
  • Advancing content provenance for a safer, more transparent AI ecosystem

    推文配文

    ⬇︎ 下载 PNG𝕏 分享到 X
#SynthID#C2PA#AI-generated images#content provenance#OpenAI
Open original article

In addition to C2PA Content Credentials, images now also contain a SynthID watermark, and can be identified using a public verification tool to check whether an image was made" / X

OpenAI on X: "We’re adding new ways for people to identify AI-generated images and understand where they came from. In addition to C2PA Content Credentials, images now also contain a SynthID watermark, and can be identified using a public verification tool to check whether an image was made" / X

Don’t miss what’s happening

Image 1: Square profile picture

OpenAI

@OpenAI

We’re adding new ways for people to identify AI-generated images and understand where they came from. In addition to C2PA Content Credentials, images now also contain a SynthID watermark, and can be identified using a public verification tool to check whether an image was made by OpenAI products.

![Image 2 Advancing content provenance for a safer, more transparent AI ecosystem](https://t.co/qo0l4vyWli)

From openai.com

5:46 PM · May 19, 2026

·

435.8K Views

264

468

3.6K

809

Read 264 replies

AI may generate inaccurate information. Please verify important content.