Google has announced a new tool for its search platform that can differentiate between AI-generated and real images to solve the problem of misinformation and fake media. The Google Search addition is called “About this image” and provides more context and information about images that you find in search results. The feature is designed to help you learn more about the images that you find, and to help you make informed decisions about whether or not to use them.
AI-generated fakes are images that are created or manipulated by artificial intelligence algorithms, such as Midjourney and Stability AI's Stable Diffusion. These AI tools can produce realistic-looking photos and videos of people, places, and events that never existed or happened. While the platforms have very practical solutions, they can also be used to create fakes.
These fakes can be used for various purposes, such as entertainment, art, education, or research. Still, they can also be used for malicious intent, such as spreading false or misleading information, impersonating someone, or altering the historical record.
Google's new tool aims to help users spot and evaluate these fakes by giving them more transparency and control over the images they encounter online. Users can access the tool by clicking three dots on an image in Google Images results, searching with an image or screenshot in Google Lens, swiping up in the Google app, or right-clicking or long-pressing on an image in Chrome on desktop and mobile.
How Google's About this Image Works
The tool will show users when the image (or similar images) was first indexed by Google, where the image may have first appeared, and where else the image has been seen online (such as news, social, or fact-checking sites).
Furthermore, About this Image will also label AI-generated images in search results, such as those created by Google itself or by other sources. For example, users will be able to see if an image was generated by Google Photos' Memories feature, which uses AI to create collages and animations from users' photos. Users will also be able to see if an image was generated by other AI models, such as Midjourney or Stable Diffusion.
Google hopes that by providing users with more information and context about the images they see online, they will be able to make more informed decisions and avoid falling for fake media. The company also hopes that by labeling its own AI-generated images, it will set a standard for other creators and platforms to follow suit and disclose their use of AI in image creation.
The tool is expected to roll out in the coming months as part of Google's efforts to combat misinformation and fake media online. About this Image is also arriving at a time when AI is becoming increasingly better at generating image and video content. Earlier this month, MidJourney 5.1 was launched with a promise of significantly improving AI art.
Generative AI is Getting Better and the Market is Growing
Midjourney CEO David Holz says that Midjourney 5.1 has a more distinctive and expressive style than the previous version, which makes it similar to Midjourney 4, a popular version for its creative and varied outputs, but with better resolution and clarity.
Midjourney 5.1 also boasts of higher precision, less unwanted edges or text glitches, and enhanced color and contrast. Users can also pick between a “raw” mode, which produces more realistic and photographic outputs, and a “default” mode, which produces more artistic and fantasy outputs. Image generation is one of the generative AI fields that has become popular recently. Google also demonstrated an image generating AI called MediaGen that is similar to other image/video AI such as MidJourney and Microsoft's Bing Image Creator.
Bing Image Creator is a new tool from Microsoft for AI and web services. The AI can make images from simple words or phrases, such as animals, objects, landscapes, or abstract concepts. But the tool also has some drawbacks and limitations. For instance, it did not make an image from the word “Bing” when it first came out. Image Creator started last month as an image-search companion for Bing Chat. It uses OpenAI and Microsoft's DALL-E image processing natural language AI model.
Last Updated on August 4, 2023 2:06 pm CEST by Markus Kasanmascheff
