Topic: c2pa standard
-
OpenAI Simplifies Image Authentication for AI-Generated Content
OpenAI has adopted the C2PA open standard for embedding metadata signals and is collaborating with Google to integrate SynthID, an invisible watermark, to improve image authentication for AI-generated content. These protective measures currently apply only to images from OpenAI's products, but th...
Read More » -
Deepfakes Are Winning the War on Reality
The proliferation of AI-generated and manipulated media is causing a crisis of trust in visual evidence online, prompting a search for solutions like content labeling. The leading technical solution, the C2PA standard for content provenance, faces critical flaws including fragile metadata and fra...
Read More » -
Olga Kornienko on C2PA, Content Authenticity & DRM at Streaming Media 2025
EZDRM's COO emphasized the collaboration between content provenance, C2PA standards, and DRM to protect media authenticity and creators from synthetic media threats. The C2PA standard establishes a chain of custody for content, verifying origin and tracking changes without judging truth, enabling...
Read More » -
OpenAI adds C2PA and Google’s SynthID to tag AI images
OpenAI has joined the C2PA open standard to attach cryptographic metadata to AI-generated images, enabling verification of their origin and edits. The company partnered with Google to embed SynthID, an invisible, machine-readable watermark that persists through cropping and compression to flag AI...
Read More » -
OpenAI adds image watermarks to help identify AI fakes
OpenAI is embedding C2PA metadata and Google DeepMind's SynthID steganographic watermarks into all images generated through its platforms, making the digital fingerprints more resilient against removal through screenshots or editing. The SynthID technology hides an invisible watermark at the pixe...
Read More » -
Instagram and X Race Against Deepfake Detection Deadline
India has mandated that social media platforms must swiftly remove illegal AI-generated content and clearly label all synthetic media, with legally binding rules taking full effect on February 20th, a move that could shape global standards. The regulations require platforms to use technical measu...
Read More » -
TikTok Cracks Down on AI Spam Accounts in Sensitive Topics
TikTok will test enhanced detection systems for AI-generated spam accounts in the sensitive categories of politics, finance, and medical content, aiming to protect public trust and well-being. The platform has already removed over 86 million fake accounts in early 2026 and has tagged over 3 billi...
Read More » -
AI Labeling Systems Face Make-or-Break Moment
Google is expanding its AI content verification tools, SynthID and C2PA, into Chrome and Search, allowing users to check images for invisible watermarks and provenance metadata from a single interface rather than separate apps. OpenAI has begun embedding SynthID watermarks into images generated b...
Read More » -
Apple will let you prove your iPhone photos aren’t deepfakes
Apple is developing "Apple Reference Image," an iOS 27 feature that embeds provenance metadata into iPhone photos at capture time, allowing users to verify images weren't AI-generated or altered. The feature, off by default, requires users to enable it in Settings and use a new "Reference" camera...
Read More » -
Oversight Board Slams Meta's Deepfake Moderation
The Meta Oversight Board criticizes the company's inadequate systems for detecting AI deepfakes, arguing its reliance on user reports and slow reviews fails to combat misinformation's rapid spread, especially during conflicts. The Board highlights a specific case of a fabricated AI video about Is...
Read More »