Chinese tech company Baidu is blocking prompts with political content for its image AI. ERNIE-ViLG is the name of the Chinese counterpart to DALL-E 2, Midjourney and Stable Diffusion. Unlike the Western AI models, ERNIE-ViLG specifically handles Chinese characters and is better with anime images. The model was trained with 145 million text-image pairs and manages ten billion parameters. By comparison, Stable Diffusion has 890 million parameters, while DALL-E 2 has a total of about 3.5 billion parameters.

While debate rages as to the merit and validity of machine-generated imagery, a photography website has fired the first shots in the man-machine war by banning images generated using AI services like Midjourney and DALL•E. PurplePort, a popular portfolio and networking website for models, photographers and imaging creatives, announced a blanket ban on "100% machine-generated images" so that the platform can remain focused on "human-generated and human-focused art". In an update titled'Artmageddon: The rise of the machines, and banning machine-generated images', owner and photographer Russ Freeman made the website's position explicitly clear. "Due to the rise of machine-generated images, we have decided to ban this type of image. Uploading images generated using services (such as Midjourney / DALL•E / Craiyon / Stable Diffusion / etc), where you type a phrase or description of the desired image and a machine algorithm (often called A.I) creates an image for you, is banned from PurplePort until further notice."

This paper presents COOL-MC, a tool that integrates state-of-the-art reinforcement learning (RL) and model checking. Specifically, the tool builds upon the OpenAI gym and the probabilistic model checker Storm. COOL-MC provides the following features: (1) a simulator to train RL policies in the OpenAI gym for Markov decision processes (MDPs) that are defined as input for Storm, (2) a new model builder for Storm, which uses callback functions to verify (neural network) RL policies, (3) formal abstractions that relate models and policies specified in OpenAI gym or Storm, and (4) algorithms to obtain bounds on the performance of so-called permissive policies. We describe the components and architecture of COOL-MC and demonstrate its features on multiple benchmark environments.

Generative AI systems create things, such as pictures, audio, writing samples and anything that can be built with computer-controlled systems like 3D printers. Discriminative systems identify things like people in pictures, words in speech or handwriting and -- most importantly -- what's real vs. The two are paired in a generative adversarial network (GAN) model. For example, a GAN for creating realistic yet fake yearbook photos might use a generative model to synthesize human faces and then pass them, along with real photos, through a discriminative model to see if it can tell which are fake and which are real. The discriminator gets better at identifying fakes, as it is told which images were created by the generator.

In pursuit of faster and more efficient AI system development, Intel, AMD and Nvidia today published a draft specification for what they refer to as a common interchange format for AI. While voluntary, the proposed "8-bit floating point (FP8)" standard, they say, has the potential to accelerate AI development by optimizing hardware memory usage and work for both AI training (i.e., engineering AI systems) and inference (running the systems). When developing an AI system, data scientists are faced with key engineering choices beyond simply collecting data to train the system. One is selecting a format to represent the weights of the system -- weights being the factors learned from the training data that influence the system's predictions. Weights are what enable a system like GPT-3 to generate whole paragraphs from a sentence-long prompt, for example, or DALL-E 2 to create photorealistic portraits from a caption.

When a demo of the software was released in late August, users quickly found that certain words--both explicit mentions of political leaders' names and words that are potentially controversial only in political contexts--were labeled as "sensitive" and blocked from generating any result. China's sophisticated system of online censorship, it seems, has extended to the latest trend in AI. It's not rare for similar AIs to limit users from generating certain types of content. DALL-E 2 prohibits sexual content, faces of public figures, or medical treatment images. The ERNIE-ViLG model is part of Wenxin, a large-scale project in natural-language processing from China's leading AI company, Baidu.

With artificial intelligence (AI) text-to-image generators exploding in popularity right now, it sometimes feels like photography is facing its most serious threat yet. In the last few months, DALL-E has been used to create shockingly realistic portraits of people who do not exist. Meanwhile, a Midjourney user even won a fine art competition using a picture he created with the software. With AI systems like DALL-E and Midjourney effortlessly churning out photo-realistic images, it may seem like there is no hope left for photography. With the havoc AI-generated art could wreck on the field of photography, for some people, this may bring to mind the way the invention of photography devastated painters in the nineteenth century.

A photographer challenged the well-known AI image generator DALL-E to recreate real-life photos that he had taken on a Leica camera, and the results are incredible, to say the least. After some prompt engineering to perfect the results from DALL-E, he presented the artworks next to one another with the text he had used to instruct the machine. The DALL-E images are marked by a multi-color band. Stelzer, a Berlin-based tech entrepreneur, shot the reference photos on a Leica M9 and a Leica M4-P and prioritized subject variety rather than aesthetics when selecting the photos to test DALL-E. "I got the idea while on vacation, and only had access to an extremely limited number of my photographs on my laptop," he tells PetaPixel.

Earlier this month, Twitter user Supercomposite posted a thread of spooky images featuring a woman she calls "Loab," who usually has red cheeks and dark, hollow eyes. Since then, the images, which range from unsettling to grotesque, have gone viral. The images of Loab all come from an artificial intelligence (A.I.) art tool. These tools, like DALL-E 2, create images based on text prompts users input into the platform--and they are having a cultural moment as of late. Just last month, a piece of A.I.-created art won the Colorado State Fair art competition.

The pace has only accelerated this year and moved firmly into the mainstream, thanks to the jaw-dropping text-to-image possibilities of DALL-E 2, Google's Imagen and Midjourney, as well as the options for computer vision applications from Microsoft's Florence and the multimodal options from Deep Mind's Gato. That turbocharged speed of development, as well as the ethical concerns around model bias that accompany it, is why one year ago, the Stanford Institute for Human-Centered AI founded the Center for Research on Foundation Models (CRFM) and published "On the Opportunities and Risks of Foundation Models" -- a report that put a name to this powerful transformation. "We coined the term'foundation models' because we felt there needed to be a name to cover the importance of this set of technologies," said Percy Liang, associate professor in computer science at Stanford University and director of the CRFM.