What happens when AI-generated images are common enough that most people can't reliably tell what's real? Google's answer, announced at I/O 2026, is to build detection directly into Chrome and Search rather than asking people to find a separate tool.
The expansion covers two distinct systems. SynthID is Google DeepMind's invisible watermarking technology - it embeds a signal into AI-generated content that's imperceptible to humans but detectable by software. C2PA (Coalition for Content Provenance and Authenticity) is an industry standard that attaches metadata to images and video recording where the content came from and what tools were used to create or edit it. Google is making both systems visible inside Chrome and Search.
In practice, when you encounter an image or video in your browser or in search results, you'd be able to see whether it carries a SynthID watermark or C2PA credentials - essentially a digital paper trail showing whether something was AI-generated.
The Coverage Gap
Neither system catches everything. SynthID only marks content generated by tools that have adopted it - a deepfake produced by an open-source model that hasn't implemented the standard won't carry any watermark. C2PA credentials only exist if the creator or platform chose to add them. Both are most reliable when content flows through platforms that have signed on to these standards.
That gap is real. But the more important shift here is distribution. A dedicated deepfake detection tool requires people to know it exists, find it, and remember to use it. Building detection into Chrome and Search means billions of users encounter these markers in ordinary browsing without any extra steps. Adobe Firefly, Canva, and other major AI image tools have started embedding C2PA credentials in their outputs. Getting those credentials surfaced in Google's products closes the loop for a meaningful portion of AI-generated content that actually circulates online.