You can now detect AI-generated videos using Gemini, but there's a catch

As the inclusion of AI in the multimedia landscape has increased so much that it is now extremely difficult to distinguish a genuine video from one generated using AI, Google's Gemini has introduced a new tool to help users identify AI-generated content.

What is Google's new AI video detection tool

But there's one major limitation: Gemini's AI-detection tool can only detect videos generated using Google's own AI models.

To check whether a video was produced through AI tools or not, users can upload a video they suspect may be AI-generated and ask: "Was this video generated using AI?" Gemini will then analyse the video for Google's proprietary SynthID watermark, which is embedded in all media produced by its AI tools.

The response will lift the veil off the concerned video's authenticity and its roots of origin. The tool will also provide details about the specific parts of the audio or video containing the watermark.

Currently, Google's this incredible AI detector supports files up to 100MB and 90 seconds in length. A similar tool to detect AI-generated images was launched last month.

Gemini's new image editing tool

Apart from that, Gemini has introduced a new markup tool for image editing, where users can upload a photo and add markings or text to specify changes. This includes adding or removing elements from particular areas. 

To implement the given changes in the final output, users must remember to click "Done".

The internet giant also released its advanced Gemini 3 model last month, which surpassed OpenAI's GPT 5.1 in various performance metrics. Following this, the ChatGPT developer responded by launching GPT 5.2 and a new image generation model to prove its competency in AI development.