Premium

Google’s experimental AI Mode can now analyse and answer questions about images

Google's AI Mode, which the company rolled out last month, can now analyse all objects in an image and answer complex questions about it.

AI Mode is currently limited to Google One AI Premium subscribers. (Image Source: Google)

Last month, Google introduced a new experimental “AI Mode” in Search, which it says lets users ask complex, multi-part queries and “get help with even your toughest questions.” For those wondering, this is similar to ChatGPT’s and Perplexity AI’s powered search results.

Now, Google has announced that it is bringing multimodal capabilities to AI Mode chatbot, enabling users to upload existing images or capture a new photo to ask questions about it. In a blog post, the tech giant said the new functionality “can understand the entire scene in an image, including the context of how objects relate to one another and their unique materials, colors, shapes and arrangements.”

Unlike traditional search queries, which only sends one query, the new AI Mode sends multiple queries about the image including the whole image and the objects in them. This means that instead of a one dimensional answer, you get more information in the generated result.

Also Read | Google DeepMind imposes tough non-competes on UK staff as AI talent wars intensify

As you see in the video, the new AI Mode with multimodal capabilities identifies and searches every book on the shelf and looks for similar books, which are then listed in the search results.

Compared to normal search queries and AI powered ones, this is much more useful since it presents users with information which would typically require multiple queries to do things like comparing options or exploring a new concept.

The new Google Search AI Mode is currently available to Google One AI Premium users living in the United States, but the tech giant may make it available to users in other countries sometime later this year.