AI Image Detector 作成者: Jordan Jones

AI Image Detector lets you instantly find out whether any image on the web was created by artificial intelligence or captured by a real camera. Right-click any image you come across while browsing and select "Detect this image" to run a deep multi-signal analysis that combines state-of-the-art transformer models with traditional forensic checks like EXIF metadata inspection, frequency domain artifact detection, and camera sensor pattern analysis. The result is a clear verdict — likely AI, likely real, or uncertain — shown alongside an AI probability percentage and a breakdown of the individual signals that contributed to the decision, so you can see exactly why the model reached its conclusion.

Beyond static images, the extension can also capture and analyze individual frames from any video playing on the page — including YouTube, LinkedIn, Facebook, Instagram, and other streaming platforms. Just right-click the video and select "Detect this video frame", or use the configurable keyboard shortcut (Ctrl+Shift+F by default) to grab the current frame and send it for analysis.

All detection results are saved to a searchable history that stays on your own device, letting you review past scans with full detail — including the original image or video frame preview, probability scores, processing time, and the complete signal breakdown. You can delete individual entries or clear the entire history at any time. The extension itself stores nothing on any server; the only data transmitted is the single image you choose to analyze, which is sent securely to a RapidAPI endpoint using your own subscription key.

The extension supports two API tiers to suit your needs and budget. The CPU tier offers fast results at a lower cost and is ideal for everyday browsing. The GPU tier uses a more powerful ensemble model for the highest accuracy with fewer false positives. You can save API keys for both tiers in the settings and switch between them at any time without re-entering your keys. A built-in test button lets you verify your configuration is working before you start detecting.

The compact popup shows the verdict, AI probability, a short summary of why the model made its decision, and the top three most influential signals at a glance. For a deeper look, open the full result page to see the scanned image preview, all signals in a table with direction indicators (AI or real), model metadata, processing time, and the complete summary. The interface adapts to your system theme or lets you choose light, dark, or system mode explicitly.

To use the extension, you will need a free or paid subscription to the AI Image Detector API on RapidAPI (see the extension settings page for links). No tracking, no analytics, no third-party cookies, and no data collection beyond the images you explicitly choose to analyze. Your API key is stored securely in your browser's built-in sync storage and is only sent to the RapidAPI endpoint you have configured.