Bing’s Chat AI Bot from Microsoft Now Allows Image-Based Searching

Microsoft has unveiled a significant update to its AI chatbot, introducing visual search capabilities. Users can now utilize Bing Chat on their desktop or through the Bing app to take or upload a photo and request more information about it.

In a release, Microsoft stated, “Bing can understand the context of an image, interpret it, and answer questions about it. Whether you’re exploring a new city and seeking information about a specific building’s architecture or you’re at home and need lunch ideas based on the contents of your fridge, you can upload the image to Bing Chat and tap into the knowledge of the web for answers.”

This update comes at a time when leading chatbot providers like Microsoft, Google, OpenAI, and Anthropic are fiercely competing in the AI race. These tech giants are rapidly launching new features to not only keep up with other text-based chatbot competitors but also to match the capabilities of image-heavy AI tools.

Although the integration of image search and image-based responses is becoming increasingly common in chatbots, none of the top text-based chatbots currently have the ability to generate their own images. This sets them apart from tools like Midjourney, Stable Diffusion, and DALL-E 2. However, Google has announced that image generation will soon be a feature of its Bard chatbot.

Microsoft’s decision to incorporate image functionality in Bing Chat follows Google’s recent introduction of an image search feature for Bard. Using Google Lens, users can request information from Bard about uploaded images, ask for captions, or simply enhance the chatbot’s responses with restaurant recommendations featuring photos of the interiors. At present, OpenAI’s ChatGPT does not support photo uploads as the chatbot remains exclusively text-based, and Anthropic’s chatbot, Claude 2, operates similarly.