Google Images -- a specialized search
engine that launched in 2001 -- has a storied history spanning decades, but to mark the feature's 25-year anniversary, the company revised its Image Search home page and added a feature
that allows users to design images from scratch inside the search results via AI Overviews.
When web results cannot find the specific visual the user is
looking for, Google allows them to build a custom graphic without leaving their browser tab.
advertisement
advertisement
An ideal image may be in the mind of a creator rather that floating across the web. The benefit
to advertisers extends far beyond the ability to pull in images from across the web, and focuses more on creating a completely new image from a person’s imagination.
If a
traditional image search fails to deliver the perfect asset, the search query acts as a creative prompt. The user can also initiate the creation of an image manually with the use of conversational
phrasing such as "help me visualize" or "create a visual."
Google brought image-generation capabilities directly into AI Overviews in Search supported by artificial
intelligence (AI) to develop this feature.
The backend runs on Google's new Nano Banana image-generation model, translating natural-language text into a new, high-quality graphic.
The update changes a simple
text prompt into a custom visual image made completely from scratch, bridging the gap between imagination and reality.
Nano Banana is Google's family of AI
image-generation and conversational editing models developed by Google DeepMind. These models are integrated into Google's platforms and allow users to create and sequentially edit "high-fidelity
visuals" using natural language, with three models.
“Image Generation” in AI Overviews will begin to roll out over the coming weeks in English, for all regions that
currently support image creation in AI Mode.
Brad Kellet, senior engineering director of Google Search, wrote in a blog post that people can create images from their imagination, but
also browse for images through a dynamic gallery as the feature pulls them in from across the web in real-time based on a person's interests.
As a user browses and saves ideas to their
collections, the images appear as tabs above the main gallery, simplifying the process to go back and continue exploring based on inspiration, similar to Pinterest.
This feature will roll
out during the coming weeks on desktop in the U.S. in English.
Both updates are built on 25 years of innovation to make the world’s visual information instantly accessible and
useful, according to Google.