In this paper, we examine how generative machine learning systems produce a new politics of visual culture. We focus on DALL-E 2 and related models as an emergent approach to image-making that operates through the cultural techniques of feature extraction and semantic compression. These techniques, we argue, are inhuman, invisual, and opaque, yet are still caught in a paradox that is ironically all too human: the consistent reproduction of whiteness as a latent feature of dominant visual culture. We use Open AI's failed efforts to 'debias' their system as a critical opening to interrogate how systems like DALL-E 2 dissolve and reconstitute politically salient human concepts like race. This example vividly illustrates the stakes of this moment of transformation, when so-called foundation models reconfigure the boundaries of visual culture and when 'doing' anti-racism means deploying quick technical fixes to mitigate personal discomfort, or more importantly, potential commercial loss.

In the last few years, the hype around artificial intelligence has been increasing (again). Most of it is due to companies like OpenAI, Google, DeepMind (Google subsidiary), Meta, and others producing truly groundbreaking research and innovative showcases in the field. From machines winning complex games like Go and Dota 2to a variety of content generation techniques that produce text, images, audio, and now video, these technologies will have an impact on our future. It feels like we have experienced this hype towards AI in the past, but it never really materialized into anything relevant to our lives. From IBM's Watson attempts to revolutionize healthcare to the prophecies of self-driving cars, we have been told about how AI will improve our society, yet there always seems to be something preventing us from getting there. On one side, technology might not be there yet for some of those advanced problems, in another, humans tend to be skeptical of machines taking over some of our areas of expertise (Skynet didn't help here).

Video game designer Jason Allen made headlines this year with Théâtre D'opéra Spatial, his submission to the Colorado State Fair's digital arts competition. Judges awarded him first place and $300 prize, but the artwork also received a sudden flurry of global attention when it was discovered Allen had used AI-powered image generator Midjourney to create the work of art. Midjourney, DALL-E and DALL-E 2 have brought a wealth of weird and wonderful images to the world as users type in natural language descriptions and share the dream-like results. DALL-E 2 uses a "diffusion model", which attempts to take the input text in its entirety and generate an image from that. But the output becomes less accurate as that text becomes more complex; the existing model appears to struggle to understand composition of concepts, and confuses attributes and relations between different objects.

OpenAI's flagship natural language processing (NLP) programme, GPT-3, has been available for two years now. Poems, essays, song lyrics, and even comprehensive manifestos could be generated by the AI language tool with just the barest of cues, and its readers were astounded by the quality of the writing produced. OpenAI's GPT-3 is a "foundation model" that was trained using what amounts to the whole of the internet (Wikipedia, Reddit, The New York Times, etc.). In order to determine the most likely responses to each given challenge, it utilises this massive information. Due to the massive scope of this study, only a select few of these basic models exist.

Shutterstock is eager to embrace AI-generated art. As The Verge reports, the photo provider has widened its deal with OpenAI to begin selling stock images built using the DALL-E 2 AI generator. The approach will offer "direct access" to DALL-E through the Shutterstock website, and compensate creators whose pictures played a role in developing the technology through a new Contributor Fund. The company also plans to pay royalties to artists when the AI uses their work. OpenAI licensed Shutterstock pictures and data to train DALL-E's text-to-image generation models in 2021.

Photo licensing service Shutterstock will begin selling images generated by artificial intelligence alongside those created by humans. The AI images will be powered exclusively by OpenAI's DALL-E 2 software. Both companies say that human creators whose work inspired the AI will be compensated, but one artist describes the move as "sewer water leak into the drinking supply". Shutterstock was one of several photo agencies that began removing AI-generated art from their archives last month. A Shutterstock spokesperson told New Scientist that the company would continue to ban people generally from uploading AI-generated art to its platform, but that its collaboration with OpenAI was an attempt to embrace new technology in an ethical way.

In recent months there has been talk about how artificial intelligence can create images from textual prompts. Therefore, when one associates the words artificial intelligence and art, one immediately thinks of DALL-E, Stable Diffusion, and other algorithms. In this article, instead, I want to discuss why artworks are often less safe than we think, and how artificial intelligence can help preserve them. "Every act of creation is first of all an act of destruction." It is a mistake to think that cultural heritage is safe. Many of humanity's most valuable works are also among the most fragile. Throughout history, only a fraction of works of art has managed to survive over time. For example, during wars, cultural heritage is often damaged.

Machine intelligence is increasingly being linked to claims about sentience, language processing, and an ability to comprehend and transform natural language into a range of stimuli. We systematically analyze the ability of DALL-E 2 to capture 8 grammatical phenomena pertaining to compositionality that are widely discussed in linguistics and pervasive in human language: binding principles and coreference, passives, word order, coordination, comparatives, negation, ellipsis, and structural ambiguity. Whereas young children routinely master these phenomena, learning systematic mappings between syntax and semantics, DALL-E 2 is unable to reliably infer meanings that are consistent with the syntax. These results challenge recent claims concerning the capacity of such systems to understand of human language. We make available the full set of test materials as a benchmark for future testing.

Join us November 12–13 for a virtual hackathon to explore how our current AI models can accelerate solutions to climate change. Learn from climate experts about the most pressing challenges, join a team through our Discord community, and push the boundaries of our language models to help address this global issue. Ambitious developers, designers, entrepreneurs and students are encouraged to apply. Please get in touch at community@openai.com if you are a startup or nonprofit working on challenging problems like climate change.

In brief Stability AI and Jasper – two startups that make AI software that auto-generates images, text, and other stuff – have each reached so-called unicorn status (valued at over $1 billion) after bagging $101 million and $125 million in funding, respectively. Stability AI, best known for open sourcing the code for its popular text-to-image Stable Diffusion model, threw a glitzy party in San Francisco this week to coincide with announcing its funding. Emad Mostaque, the company's founder, took to the stage announcing plans to build and release more AI tools capable of handling text, audio, and video. Stability's latest funding round was led by Coatue, Lightspeed Venture Partners, and O'Shaughnessy Ventures LLC. A day later Jasper, a startup that uses OpenAI's GPT-3 to output text and images, also announced a successful Series A round.