AI is available to Google Cloud Vertex customers; new generation brings more realistic “photos” and can show Bard’s future
Google announced this Wednesday (13) Imagen 2, the new (and second) generation of its AI model for creating images. This artificial intelligence is only available to Vortex AI customers, a service available from Google Cloud. In previews shown by big tech, Imagen 2 delivers ultra-realistic results.
Google also claims that Imagen 2 has improved its understanding of prompts, which should lead the user to achieve the desired results (or as close to it as possible). Unfortunately, as it is a paid service within another paid service, we cannot test it and there is no testing space available from the company — and the Gemini case leaves us behind.
Images 2 trained with reading subtitles
In the official announcement of Imagen 2, made on the DeepMind website, a subsidiary of Google, the company reveals that to improve the understanding of the prompts , worked on expanding the description of images used in AI training. According to Google, this allowed Imagen 2 to improve understanding of the context and nuances of commands.
The AI model used in the tool also received “specialized training” in aesthetics based on human preferences. Google created an aesthetic scale (the popular Instagrammable) for each image, taking into account factors such as lighting, exposure time, framing and sharpness. The company claims this allows Imagen 2 to generate even higher quality images.
This ability, added to the example given by Google, raises a question: how long will AIs be present in social media feeds through beautiful Instagrammable images? Images of people are becoming more and more realistic, but why would anyone be interested in publishing a photo of another person or someone similar? Images of beautiful landscapes are still more attractive for likes.
Imagen has a tool to recognize your images
A differentiator of Imagen’s AI is that it has a watermark that recognizes images created by itself. This Google technology creates an identifier in the pixels of photos and artwork created by artificial intelligence and, according to the company, cannot be removed in any way. Thus, a user can upload an image to Imagen and it will identify whether it was created by AI or not.
The negative point for Imagen is that it remains restricted to Google customers. Bard is not yet capable of generating images (although he says yes, he just searches for them on Google). Googlegenerative AI uses LLM Gemini, but Google’s plans to bring image creation to Bard are unknown. However, Imagen already gives a preview of what we can see in the future (distant or near).
