This week marks the 25th anniversary of the launch of Google Images, the feature that brought a visual component to Google Search thanks to the flood of queries from users who wanted to see the iconic green Versace dress Jennifer Lopez wore to the 42nd Grammy Awards ceremony, which I am sure you know well (whether you lived through it, know the story, or have watched Off Campus recently).
To commemorate the anniversary, Google has unveiled two new updates related to visual search. The first is tied to the generative capabilities of AI Overviews, and the second is a new homepage for Google Images.
Google has integrated the capabilities of Nano Banana into its AI Overviews, allowing users to generate images directly in Search. All they need to do is type a prompt into the search bar describing the image they want to create, and the AI will generate it along with a text description that includes follow-up suggestions to refine it further.
“This update transforms a simple text prompt into a high-quality custom image, created entirely from scratch, seamlessly bridging imagination and reality,” the company explains.
The feature will begin rolling out over the coming weeks in English, in those regions that already support image creation in AI Mode.
That said, I have to ask: is it really necessary to bring generative AI to every user touchpoint? My short answer would be no. But since we are here, let me expand on that a bit.
Generative AI image tools are relatively new compared with many other technologies, and even so, we are already seeing how content created with them has started to flood the internet. Fully aware both of how useful these AIs can be and of the problems they pose for real artists (and of the fact that the former cannot match the value and quality of a work created by a real person), my main concern here is how dangerous it is when reality starts to blur.
For example, I can search for information about fashion in the Victorian era and receive real documentation in response, such as images from newspapers of the time, alongside designs created by professionals who have researched and trained in the subject. That simply is not comparable to an AI-generated image that may contain inaccuracies, whether due to licensing issues, design errors, or because it mixes references from other styles. In the same way, searching for an image of a cheetah and getting a real photograph will always be more accurate—anatomically, in terms of how its body actually moves, its habitat, and so on—than having Nano Banana generate the result, no matter how advanced its capabilities have become.
Most likely, this is not quite where Google is aiming, and what the company really wants is for users to use this new feature for tasks related to bringing their ideas to life. As Google itself put it, “Sometimes, the perfect image is out there on the web, waiting to be discovered. But other times, you may have a very specific vision for which no image exists yet.”
In fact, the example shared on its blog showed the result of the request “create an image comparing a small bathroom with dark green walls and butter yellow walls.” The AI Overview generated the image and added text with suggestions and questions to further refine the design if the user wanted to do so. Something that could just as easily have been done in the Gemini app and that, in my view, does not add much value to Search either.
I understand the logic of getting users to sample AI features across the entire ecosystem in order to build habit and eventually create a sense of need around having access to them, but taking a moment to reflect on how this technology is evolving and where it is leading us is never a bad idea.
The second update announced by the tech giant is a new homepage for Google Images. It features a dynamic, immersive gallery of images from across the web, updated in real time and personalized based on your interests.
“As you explore and save ideas to your collections, they will appear as tabs above the main gallery, making it easy to revisit and explore based on what inspires you,” Google explains.
This new page will begin rolling out on desktop in the United States and in English over the coming weeks. You need to be signed in to your Google account to use it.
Photo: Google and Canva
Your email address will not be published. Required fields are marked *
Δ