Google Brings C2PA to Gemini App, OpenAI Adds SynthID to AI Images as Industry Pushes for Transparency

The artificial intelligence (AI) industry is slowly adopting a multi-layered content verification system to help users understand when content is digitally created or altered, and when it is created organically. At Google I/O 2026, the Mountain View-based tech giant announced integration of the Coalition for Content Provenance and Authenticity (C2PA) content credentials to the Gemini app. After Pixel devices, this is the first time the verification system is being used in the tech giant's products. At the same time, OpenAI announced that it has partnered with Google to bring SynthID watermarking to its AI-generated images.

Google, OpenAI Strengthen AI Content Verification

Now that AI models can generate text, images, audio, video, and 3D models, they have started blurring the line between what's real and what's artificial. The problem, also known as deepfakes, is both a cybersecurity challenge and a social nightmare. The industry, as a whole, has been finding ways to deal with this problem ever since the first foundational models started appearing.

One of the earliest solutions focused on multimedia content and offered a universal protocol to embed the AI information into the metadata of the file. The protocol was called C2PA, and many AI companies, including OpenAI, Microsoft, Meta, and Google, adopted it. At the same time, Google also developed a watermarking technology called SynthID, which is embedded at a pixel level to ensure no amount of editing or cropping can remove it. There were other technologies as well, but none gained enough mainstream attention.

However, there was a problem. Different companies leaned towards different solutions. While Google focused on creating SynthID for different output formats, OpenAI became a big proponent of C2PA. Fragmented adoption also resulted in a lack of universal detection systems that can quickly verify whether an image was generated using AI or was captured using a camera.

This seems to be changing now. At I/O 2026, Google finally announced the adoption of C2PA for the Gemini app, highlighting that it will also be added to Search and Chrome in the coming months. Similarly, OpenAI also announced that it is now adopting a “multi-layered, ecosystem-driven model” by adding SynthID watermarking to images generated by its large language models (LLMs).

“These two systems reinforce each other. C2PA helps content carry detailed context; SynthID helps preserve a signal when metadata does not survive. Watermarking can be more durable through transformations like screenshots, while metadata can provide more information than a watermark alone. Together, they make provenance more resilient than either layer would be on its own,” the company stated.