Category
Research
7 articles in this category.
What Is SynthID & Can It Be Removed?
Most people picture a watermark as something you can see, or at worst something written into a file's metadata. SynthID is neither. It is a signal baked into the content itself at the moment of gener…
Which AI Images Are Easiest to Detect?
"Can detectors tell?" is the wrong question. Two images from two generators, of the same subject, at the same resolution, can sit at opposite ends of the detectability scale — and most of the gap has…
Does Google Penalize AI Content?
"Will Google penalize my AI content?" is one of the most anxious questions in SEO right now, and it's usually asked with the wrong model in mind. People imagine Google running an AI detector over eve…
The AI Detector Landscape in 2026
The detector market in 2026 looks nothing like the detector market in 2023. Three years ago, "AI detection" meant a single confidence score from a single model. Today it means a fragmented ecosystem…
Why AI Video Detectors Fail in 2026
In 2024, AI video detectors worked well enough to be useful. A trained classifier could spot Sora 1 and Runway Gen 2 output with around 90% accuracy on standard benchmarks, and the public tools — Hiv…
C2PA & Content Credentials Explained
Content Credentials are the most consequential thing to happen to image authenticity since EXIF, and one of the most widely misunderstood standards on the open web. The press coverage tends to oscill…
PRNU, FFT & Sensor Noise Explained
Image forensics is a small, mathematically dense field that quietly underpins everything from courtroom exhibits to AI detection startups. Three pillars do most of the heavy lifting: PRNU (the sensor…