Loading page...Загружаем страницу...正在加载页面...
AtlasRepo

Is SynthID reliable? I tested Google's AI image check on screenshots, stripped files and real photos

Описание и применение

Тест показал, что водяной знак SynthID Google обнаруживает AI‑изображения даже после удаления метаданных и при скриншотах, но пропускает небольшие вставки и сайт Detector был недоступен.

Задача

Determine whether Google’s SynthID watermark detection works on screenshots, metadata‑removed files, and real photos.

Результат

SynthID reliably flags AI‑generated images even after metadata removal and when captured as a screenshot, but it missed a small pasted AI segment and the public Detector site was down during the test. Real photos without AI content were correctly identified as non‑AI.

Рабочий процесс

  • Generate two Gemini images (2400×1792)
  • Create test variants: original, metadata‑removed, screenshot PNG, NASA photo, NASA photo with sky edited, AI image pasted into photo
  • Attempt uploads to SynthID Detector (failed with error)
  • Use Gemini app’s AI‑generation check for each file
  • Record Gemini’s responses and content credentials presence

Доказательства

Google's SynthID check in the Gemini app flagged a Gemini-generated image after its metadata was removed and again after a screenshot, and found nothing on a real photo.

Источник

The Detector was the plan, but on October 12 all four of my uploads, between 4:36 and 5:33 pm UTC, ended in "Something went wrong! Please wait and try again later."

Источник

Gemini said about each file: ... "Most or all of this image was edited or generated with Google AI." (2 of 2 runs)

Источник

  • Дата записи: 2026-10-12T19:56:13.180Z