Trendora
vien

C2PA

Trial

Techniques

A content provenance standard used to sign supported image and file outputs with verifiable metadata.

Why it's here

Placed in Trial: 7 article(s) of evidence from 5 source(s), led by product launches, with 4 in the last 30 days. Confidence 76%.

Evidence (7)

  • 7Ars Technica AI·10/6/2026regulation
    OpenAI to watermark ChatGPT text by default in the EU

    OpenAI says it will automatically watermark text generated by ChatGPT in the European Union to align with the EU AI Act. Outside the EU, the feature will still be available but turned off by default, and the company plans limited access to its detector for selected researchers and organizations.

  • 8The New Stack·9/24/2026product_launch
    Google adds self-serve custom voice creation to Gemini TTS

    Google has released Gemini 3.8 Flash TTS and Flash-Lite TTS with new self-serve voice creation features in the Gemini API and Google AI Studio. Developers can now design voices from descriptions or replicate a speaker’s voice with consent, save reusable voice IDs, and use content credentials and SynthID to mark generated audio.

  • 7Hacker News·9/16/2026product_launch
    Apple unveils Reference Image for verifiable iPhone photography

    Apple describes Apple Reference Image, an opt-in camera mode for iPhone 18 Pro and iPhone 18 Pro Max that creates a securely timestamped reference image intended to prove a photo was captured by a real sensor at a specific time. The system combines dedicated secure hardware and Private Cloud Compute to preserve image integrity and privacy, and it can revoke fraudulent images if compromise is detected.

  • 5Hacker News·9/10/2026open_source
    Open-source camera proves capture with steganographic signatures

    A Recurse Center project built a low-cost camera that signs images at capture time using a secure element and hides the signature inside the pixels. The system uses a perceptual hash and watermarking so the proof can survive compression and resizing, though it still cannot stop screen rephotography attacks.

  • 5Hacker News·9/2/2026product_launch
    Claude file provenance checker

    Anthropic has introduced a browser-based tool that checks whether a supported file contains a Claude content credential, indicating it was made or processed with Claude. The checker runs locally in the browser and reads only embedded C2PA provenance metadata, without uploading the file or identifying the user.

  • 7Hacker News·8/24/2026research
    Microsoft Paint and Photos Embed Invisible GUID Watermarks in Local AI Images

    Reverse engineering shows that Microsoft Paint and Photos send prompts to a remote moderation server, which returns a GUID that is then embedded as an invisible watermark in locally generated images. The visible watermark toggle does not control this behavior, and Microsoft also uses C2PA metadata for AI-generated image saves in supported formats.

  • 7Anthropic News·8/11/2026regulation
    Anthropic adds invisible watermarks to Claude text

    Anthropic said new Claude models will embed invisible, machine-readable watermarks in generated text across its API, coding tools, and supported cloud platforms. The move is meant to help trace AI-generated output and align with EU AI Act transparency rules, but Anthropic says the watermark is not strong enough to prove authorship and may not work in every workflow.