Google is rolling out visual search support inside its AI Mode, giving users the ability to upload or snap a photo and ask questions based on what they see.
AI Mode is not the same as the AI Overview you might have seen at the top of search results. This is a separate, dedicated interface powered by Gemini 2.0, designed for open-ended conversations.
Unlike AI Overview, it supports follow-up questions and includes citations for answers. Think of it as a blend between classic Google search and tools like Perplexity.
The new visual search feature combines Gemini’s language abilities with Google Lens. You can show the AI a photo of books, a meal, a room, or any object. Lens identifies what is in the image and sends that data to Gemini, which breaks it into multiple sub-queries. The result is a more personalized and layered answer.
For example, if you upload a photo of several books, AI Mode might tell you which one is the most popular, offer summaries, and suggest similar titles based on your follow-up.
AI Mode was previously limited to paying Google One AI Premium subscribers. Now, it is being expanded to free users in the United States through Google Labs. You still need to opt in, but the feature is no longer locked behind a subscription.
