Google says its tools helped create 100 billion AI images. A watermark can label them, but can it stop the harm?

Key points

• Starling Lab, a research collaboration from Stanford University and the University of Southern California, estimates humanity took 149 years to produce 1.5 billion photographs; generative AI matched that in 18 months.
• Google announced at its I/O developer conference this spring that its tools helped create more than 100 billion AI images and videos in just a couple of years.
• Google is expanding partnerships around SynthID, its technology for tagging AI-generated content, so more platforms can use it.
• SynthID can label AI content at creation, but the label cannot stop fake images from spreading before anyone checks.

Put a number on it: humanity snapped photographs for 149 years before reaching 1.5 billion images. Generative AI, software that creates photos, videos and audio from a text prompt, crossed that same line in 18 months. Those figures come from Starling Lab, a research collaboration run jointly by Stanford University and the University of Southern California that studies how digital media is authenticated.

Then Google arrived with a bigger number. At its I/O conference this spring, the company said its tools had been used to produce more than 100 billion AI images and videos over just a couple of years. First reported by Ars Technica AI, that figure covers Google's own products alone. Dozens of other companies also run image and video generators.

What is SynthID, and how does it work?

SynthID is Google's answer to a simple question: once an AI image exists, how does anyone know it is fake? The tool bakes an invisible watermark, a hidden signal woven into the pixels or audio data, into content the moment Google's systems create it. A separate detector can then read that signal later.

The watermark survives a surprising amount of abuse. Cropping, colour filters and screenshot re-saves generally leave it intact.

Google used the I/O announcement to confirm new partnerships that would spread SynthID to more platforms, meaning more AI content created elsewhere could carry the tag.

So is the problem solved?

Not even close. A watermark is a label, not a lock.

Content that carries a SynthID tag can still spread across social media, messaging apps and news feeds long before any detector scans it. Most ordinary people have no way to run a check themselves, and platforms are not required to display a warning even if they could.

There is also the content that never gets tagged at all. SynthID only marks what Google's own tools produce. Any image made with a rival generator, or with older open-source software that anyone can download and run locally, carries no watermark whatsoever.

What should readers watch for?

Labelling technology is genuinely useful and worth having. But volume is the real story here. When any single company's tools can produce 100 billion pieces of synthetic media in two years, the realistic risk is not one convincing deepfake; it is an ocean of plausible-looking fake content that overwhelms the human capacity to question it.

A few practical habits help. Reverse-image search any photo that seems designed to provoke a strong reaction. Check whether a story appears on more than one credible outlet before sharing it. If a video shows someone saying something shocking, look for a text transcript from a named journalist before trusting it.

The watermark is a starting point. Scepticism is still the better tool.