A study by the Stanford Internet Observatory has revealed that popular AI image generators, including Stable Diffusion, have been trained on thousands of images of child sexual abuse. The research identified over 3,200 suspected images of child sexual abuse in the LAION database, prompting LAION to temporarily remove its datasets. The presence of these illegal images is believed to be influencing the generation of harmful outputs by AI tools and contributing to the creation of explicit imagery of fake children. The Stanford Internet Observatory is calling for drastic measures, including the deletion of training sets based on LAION and the removal of older versions of Stable Diffusion from the internet.
Meta AI's Imagine, a free AI image generation tool, competes with Microsoft's Bing Image Creator and Stable Diffusion. While Bing generally produces the best images, Imagine offers advantages in terms of speed and image size. However, Imagine is more reluctant to generate images without people and requires users to log in with a Meta, Facebook, or Instagram account. Setting up a local install with a capable GPU provides complete control over image generation parameters and avoids content filters.
AI image generators have become essential tools for creators in the digital art and design world. This article compares and analyzes leading AI image generators, including MidJourney, DALL-E 3, Stable Diffusion, Leonardo AI, Adobe Firefly, and Amazon Titan. Each tool has its own strengths and weaknesses, catering to different user preferences and needs. Factors such as image quality, ease of use, text generation capabilities, customization options, and cost are considered. Users are advised to choose the tool that aligns with their creative vision and workflow.
NVIDIA has released a new Game Ready driver that enhances performance for DLSS 3 in NARAKA: BLADEPOINT and Warhammer: Vermintide 2. The driver also includes updates to RTX Video Super Resolution, improving image quality and expanding support to GeForce RTX 20 Series GPUs. Additionally, Stable Diffusion performance is now up to 2X faster with TensorRT acceleration. The driver also adds support for optimal settings in over 14 new games. Users can download the driver through GeForce Experience or GeForce.com.
MidJourney, a generative AI image creation tool, has introduced a new feature called "Vary (Region)" that allows users to engage in inpainting, editing specific portions of an image using text prompts. While this marks a significant development for MidJourney, it faces competition from Stable Diffusion, an open-source image generator that offers similar functionalities with better accuracy and user experience. Despite its limitations, MidJourney's Vary tool showcases the ongoing advancements in AI image editing, promising a future of enhanced creativity and customization.
Digital artist Greg Rutkowski, known for his vivid style, opposed the use of artificial intelligence (AI) in art. In response, AI art creators removed his work from the popular AI image generator Stable Diffusion. However, the art community created a tool to mimic Rutkowski's style using Stable Diffusion's open-source code. Despite the controversy, the move highlights the ongoing debate between innovation and infringement in the evolving world of AI and art.
Stability AI has announced that the next version of its prompt-based AI image generator, Stable Diffusion, called SDXL 0.9, will produce more photorealistic images and better hands. The new model provides a significant increase in parameter count, allowing for improved results. It will soon be available on Stability AI’s Clipdrop web tool and will be added to the startup’s DreamStudio app. The open-source version of SDXL 1.0 will arrive in mid-July.
Amsterdam-based oio's lead designer, Bjørn Karmann, has created Paragraphica, a device that produces photos without a lens or sensor. Instead, it relies on a web application to collect data points, which it interprets and converts into an AI-generated image using Stable Diffusion. The device uses a Raspberry Pi 4 Model B and a 15-inch display. While there are no plans to mass-produce the device, the software can be tested on Karmann's website.
Stability AI is releasing an open-source version of DreamStudio, its commercial interface for the Stable Diffusion AI image generator model, called StableStudio. The move is part of the company's dedication to advancing open-source development and community building. Stability AI has previously released various versions of Stable Diffusion for free download and last month released a suite of open-source large language models. However, the company's approach sometimes lacks direction, with StableStudio potentially competing with DreamStudio, which the company plans to customize for corporate clients.
