New integration allows users to create images from text prompts in Google Search. While innovative, questions arise around safeguards against misuse of this groundbreaking tech.
AI Visual Generation Comes to Google Search
Google has integrated an AI image generator feature directly into their search engine, marking their entry into AI-created visual content. Using Google Imagen, still in beta testing, users can now generate images from text prompts without needing access to Imagen itself. Until now, only select users could fully utilize Imagen through Google’s AI Test Kitchen. For others, access was limited to Imagen-powered specialty generators.
The new integration makes Imagen’s capabilities available to a wider audience through Google Search. It is initially only offered in English to U.S. users who opt into the Google Search Generative Experience. Users can seamlessly create images from text by opting in and using the feature directly in search. It will also become an option when browsing image results, allowing image generation if desired visuals aren’t found.
Users can specify if they want their image as a photo, drawing or painting. Google will provide four tailored options to choose from before downloading the preferred image as a PNG file. This is similar to Bing’s integration of DALL-E for image generation, but Google promises strict content filtering to prevent offensive content.
Safeguards and Limitations Established to Prevent Misuse
To maintain community standards, Google has set safeguards to prevent misuse. These include blocking photorealistic faces and restricting names of famous people in prompts. Initially, only users over 18 will have access, and all images will have watermarks and metadata identifying them as AI-generated.
However, doubts persist about the effectiveness of these safeguards. Bing faced similar challenges, with users circumventing protections. Instances of people bypassing safeguards on other AI features, like Meta’s AI stickers, further highlight the issue of misuse.
Continuous Enhancement Through User Feedback
Google acknowledges its AI generator may have flaws. This is why an opt-in approach was chosen. To improve the tool and tackle problems, users can provide feedback on any shortcomings they encounter. Unlike Bing, Google hasn’t indicated if they will charge for the feature. If you want you can also read- Adobe’s New AI Photo Editing Tool Could Be the Future of Photo Editing
More Generative Capabilities Added to Google Search
Beyond image generation, Google Search Generative Experience introduces other features. One is conversational search powered by integration with Bard, similar to Bing and ChatGPT. Users can also draft customized text content, adjusting length and tone of generated output. These are currently available in the U.S. to those who opt into Google Search Labs.
Google’s inclusion of AI image generation signifies major progress in AI content creation. But its success depends on maintaining strong safeguards and addressing risks. As this innovative feature evolves, it opens new possibilities for enhancing visual content creation, if used responsibly.
