Google is experimenting with incorporating video into the search process. This initiative, powered by their innovative AI model Gemini, allows users to search using a combination of uploaded videos and text queries.
Imagine encountering an unfamiliar object or malfunctioning appliance. Previously, describing the issue through text alone might not yield the most accurate results. With this new feature, users can upload a short video showcasing the problem and ask a question. Gemini then analyzes the video frame-by-frame, identifying the object or situation. It subsequently employs its vast knowledge base to deliver an “AI overview” with potential solutions or troubleshooting tips. Additionally, users can access links to relevant discussion forums or even instructional videos for a deeper dive.
Source: Google
This development comes at a time when video platforms like TikTok and Instagram are gaining immense popularity, especially among younger users. Google acknowledges this shift and views integrating video search as a way to cater to evolving user preferences. This feature builds upon Google’s existing multimodal search capabilities, which already allow users to combine text with images for enhanced search results.
Source: Google
The initial launch will be a limited experiment within Google Search Labs, accessible only to users in the United States searching in English. This controlled rollout allows Google to gather valuable user feedback and refine the technology before potentially expanding its reach.
The implications of video search are significant. It has the potential to improve search accuracy, particularly for queries with a visual component. Additionally, it caters to users who might find the video a more natural or efficient way to express their search intent compared to text alone.
