A new feature has appeared in the chat with Alice AI – the neural network can now combine two images into one. The user only needs to upload photos and describe the result, after which the system will independently create a new image.
In fact, this simplifies the creation of visual content: you no longer need complex graphic editors or shooting. Everything can be done directly in the chat – from creative experiments to work tasks.
For example, now you can "take a photo" with a famous person – find yourself at a table with Keanu Reeves or take a walk in the park with Margot Robbie. The function also allows you to create avatars, try on clothes or change your appearance – just upload your photo and the desired style.
The practical application is also wide: the neural network can generate images for product cards, websites or advertising. For example, show clothes on a model or "place" furniture in different interiors – without photo shoots and studios.
The process is as simple as possible: in the chat, you need to select the "Combine photo" function, upload images and describe what should happen. The neural network analyzes objects, lighting and composition, and then, using a diffusion model, assembles the final picture, maintaining realism and recognition.
All tools for working with images are collected in the "Studio" section. Previously, photo animation and image editing functions had already appeared in the service – according to the company, users have processed hundreds of millions of images.
At the same time, the development of such technologies is taking place against the background of tightening regulation. Measures of responsibility for the creation and distribution of deepfakes are already being discussed in Russia: the corresponding bill provides for penalties for abuse of such technologies.
Against this background, the new capabilities of neural networks are attracting more and more attention – the line between a real image and a generated one is becoming less and less noticeable.