On December 20th, according to the tech-media Phone Arena todayGoogle Gemini New AI VideoTesting functionHELPS USERS TO JUDGE WHETHER THE VIDEO IS REAL OR AI GENERATED.
In conjunction with an earlier report by 1AI, Gemini has been able to test AI images in November of this year, and now it is being further expanded to video。
Using Gemini to test the AI video is very simple. Users need only click on the cell phone to access Gemini App or access the web page on the computer, then upload the suspected AI-generated video and enter similar "Is this video created by AI?" And after a while, Gemini will give the answer。
It should be noted, however, that Gemini is currently only able to identify the video generated by Google's own AI tool, and that the file size cannot exceed 100 MB and the video cannot exceed 90 seconds。
In principle, Google uses a technique called SynthID to mark digital watermarks of content generated by all of its own AI tools, which cannot be detected by humans, but can be easily identified by Google ' s own detectors, while other home-generated AI elements such as ChatGPT and Midjourney are not embedded in the SynthID watermark, so Gemini cannot be identified。