Google has introduced a significant update to its AI Mode, enhancing users’ ability to search for images using imprecise descriptions and references. The new functionality aims to create a more conversational style when shopping online, allowing users to express their needs more naturally, according to the company’s announcement.
In their announcement, Google stated, “Now, you can pose a question in a conversational manner and receive a variety of visual options in AI Mode, with opportunities to continually refine your search in the most intuitive way for you.” Users can expect to see visually rich results that align with their preferences and can follow up by requesting more variations, such as “I want more options with dark tones and bold prints.”
For instance, a user could search for “barrel jeans that aren’t too baggy” and then further refine the results with additional prompts like “I want more ankle length” or “show me acid-washed denim.” Google has indicated that AI Mode will intelligently suggest relevant, shoppable options, making it simple for users to navigate to retailers’ sites directly from the search results.
Users now have the additional option to initiate searches by uploading a reference image or taking a snapshot to receive visually similar results. This functionality can also be combined with written descriptions for a more detailed search experience. Furthermore, the updated conversational search features are applicable for general visual exploration, such as seeking inspiration for interior design.
The AI Mode upgrade for visual search is being rolled out in English to users in the United States this week, although it may take several days for the full suite of new features to become available. Google noted that this enhancement expands upon the existing Google Search with Lens and Image search capabilities, utilizing Gemini 2.5’s advanced multimodal and linguistic functions. This allows the AI Mode to identify finer details and secondary objects in images, leading to a more comprehensive understanding of visual context and improved image search outcomes.
