Born from the ingenuity of tech entrepreneurs Elon Musk, Greg Brockman, Ilya Sutskever, Wojciech Zaremba and John Schulman, DALL-E is an augmented reality algorithm capable of generating images from textual descriptions.

The meltdown of art history may have started with a cleavage - specifically that of Jennifer Lopez in the <a href="https://www.lofficielibiza.com/fashion/gianni-versace-assassination-history-career" target="_blank" rel="noopener">Versace</a> dress worn at the 42nd <a href="https://www.lofficielibiza.com/music/2023-grammy-award-nominations" target="_blank" rel="noopener">Grammy Awards</a> on February 23, 2000. Eager to check it out, hordes of fans googled "J.Lo Grammy Dress" but found nothing because image search had yet to be invented. The sheer number of searches for that dress inspired Google to create Google Images , which they rolled out in July 2001. Twenty-one years later, the pioneering concept of publicly available, searchable photos has amassed more than 1.12 billion images for the Internet. In the breadth of data that we have uploaded, shared, created and forgotten, we have created a new formula for the accumulation of the image and, without knowing it, we have developed a vast pool of data for the training of <a href="https://www.lofficielibiza.com/art/dall-e-2-ai-digital-art-images" target="_blank" rel="noopener">AI researchers - artificial intelligence</a> - and the algorithms to use. It took years to get to the current processing power of text-to-image generators. For the most part, the image labeling work was the most significant contribution, requiring an exceptional amount of man-hours. In 2007, computer scientists at Stanford and Princeton began classifying images derailed by the fast labels that general users simply attached to their "cat jpeg" photos. They realized the need for human-generated input to create sophisticated image titles for classification programs that can optimize search power. (That's the difference between "cat" and "big fluffy white cat in a basket"). The project has employed more than 20,000 people who have tagged images; by 2012, the team had built a database of 14 million titled photos. And once satisfied with the progress of the work, the scientists hung a poster advertising the data collection at a Miami Beach conference center, which quickly evolved into an annual competition to see which algorithms could identify objects in the data collection images. with the lowest error rate.
Scroll down to find out more about the DALL-E augmented reality algorithm

"Tech CEOs imposing their will unto you art world in the style of Cindy Sherman" BY ALEXIS SCHWARZ

Thus began the genesis of a generative adversarial network, known as GAN - Generative Adversarial Network. GAN started out as a battle of the geeks. As <a href="https://www.lofficielibiza.com/hommes/loewe-fashion-show-man-spring-summer-2023-collection-photo-look" target="_blank" rel="noopener">technology</a> writer Loz Blain explains, GANs put it: "Two artificial intelligences competing with each other, both 'trained' because huge amounts of real images were shown, titled to help the algorithms learn what were they to look for. An "AI generator" then starts creating images and a "discriminator " Al tries to guess whether they are real images or AI creations". After billions of iterations, AI has become more sophisticated at creating and deciphering, but it remained rudimentary in its imaginative conceptual reasonings. At times, AI was convinced of the reality of photos that children would consider unrealistic. In 2016, Andrew Conru, a businessman with a Ph.D in mechanical engineering at the Standford University, which had helped develop then-pioneering shopping cart and online matchmaking services, took GAN a step further. For Conru, "The logical next step on the path of artistic evolution" was to outsource the to engineers who, with a vast understanding of hardware and software, could make painters obsolete. From there, RobotArt was founded. The competition, powered by the Smithsonian and Popular Mechanics, was translated into a dystopian "nerdgasm" of robotics enthusiasts superficially recreating paintings without the human hand. Some robots created "new works", others replicated <a href="https://www.lofficielibiza.com/art/marisol-sol-escobar-andy-warhol-pop-art-legacy-exhibition" target="_blank" rel="noopener">Pop art</a> , while others were coded to stylistically create "new" versions of Van Gogh's impasto via AI generators. In the end, the winners shared a $100,000 prize, packed up their robots and went home, unfazed by the metaphorical slap. But then came GANism. Because artists emerge within a space to free the medium from its commercial chains.
Scroll down to find out more about the DALL-E augmented reality algorithm.

"The courage to be real" BY STUART VEVERS

Artists like <a href="https://paglen.studio/" target="_blank" rel="noopener">Trevor Paglen</a> and researcher Kate Crawford have rewritten labels and fed them through their own patented system to create new data. Parisian collective Obvious created a GAN system using portraits painted between the 14th and 20th centuries, then generating an AI portrait sold at Christie's for $432,500. Meanwhile &ldquo;Deep Meditations&rdquo;, the 2018 film by Memo Akten, trained on images tagged with abstractionist concepts that described the meaning of existence. The medium caught on, encouraged by the critical methodology approaches of Casey Reas, Lev Manovich, Arthur I. Miller, and Nora N. Khan, interested buyers, and innovative approaches to deconstructing the GAN. GAN Art superstars such as <a href="https://www.lofficielibiza.com/design/exhibition-bulgari-snakes-metamorphosis-saatchi-gallery-london" target="_blank" rel="noopener">Refik Anadol</a> and Mario Klingemann and alumni of Google's Arts &amp; Culture residency began to emerge, while institutions, including MoMA and the Center Pompidou, flung their doors wide open. The art world has finally found a lens through which GAN can make sense, primarily as a subversion of their Valley origins. Artist-created image titles have become a nuanced approach to examining our systems of life and perception. In July 2022, another GAN iteration was released: its name is DALL-E 2. A brilliant mix between Pixar's WALL-E and Salvador Dali, it may be one of the most extraordinary and potentially apocalyptic technologies ever created. The brainchild of tech entrepreneurs Elon Musk, Greg Brockman, Ilya Sutskever, Wojciech Zaremba and John Schulman, Open AI - the parent company of DALL-E - has created generative capabilities tailor-made for the art world Without precedents. Are you interested in an &ldquo; <a href="https://www.instagram.com/annieleibovitz/?hl=en" target="_blank" rel="noopener">Annie Leibovitz</a> -style Alien on Kodak 400 film&rdquo;? DALL-E 2 has it.
&ldquo;The Supreme Court is rocking the floor Norman Rockwell style&rdquo; ? Yes sir, <a href="https://www.lofficielibiza.com/art/dall-e-2-ai-digital-art-images" target="_blank" rel="noopener">DALL-E 2</a> can do that. There has never been a text-to-image generator for the art world before, and never before has anyone been adopted so quickly. In late summer 2022, OpenAI sent out one million DALL-E 2 beta invitations to insiders, including artists, journalists, researchers, entrepreneurs, and taste-makers. And within a week DALL-E 2 was ubiquitous on Instagram, in group chats, and on Twitter. Painters and tech entrepreneurs alike have adopted the generator: Cover-star Emma Stern made posts, and billionaire LinkedIn co-founder Reid Hoffman sold his DALL-E 2 creations for $24,000 on <a href="https://www.lofficielibiza.com/fashion/givenchy-launches-a-capsule-collection-street-swear-associated-with-nfts" target="_blank" rel="noopener">NFT marketplace,</a> Magic Eden. Using the Fyre Festival pattern (the famous orange square), DALL-E 2 has become cool. But, like the ill-fated festival, it's important to note that the users are not the benefactors and that management is in crisis. OpenAI made a metamorphosis with respect to its original statute, where the company's claim was: &laquo;The first fiduciary duty is towards humanity&raquo; , when it developed AGI - a machine that has learning and reasoning powers typical of the human mind, on - perishing the line of singularity. AGI technology, feared by those who understand its existential threat, would be less palatable if it were sponsored by the greed of big-money corporates, so to start, OpenAI was registered as a non-profit.

As Karen Hai wrote in MIT Technology Review: &ldquo;Although no criticism was ever made, the implication was clear: other labs, like DeepMind, may not serve humanity because they are driven by commercial interests. While those are closed, OperAI could be open&raquo; . Or so we thought. Flash forward just a few years, OpenAI revoked its nonprofit charter and registered as a business in 2019 &mdash; a controversial decision and a pivotal choice. Soon after the announcement, <a href="https://www.lofficielibiza.com/culture/celebrities-leaving-twitter" target="_blank" rel="noopener">Elon Musk</a> , who believes AI is: &laquo;Potentially more dangerous than the atomic bomb&raquo; , exited OpenAI with a tweet: &laquo;I disagreed with some of the things the OpenAI team wanted to do&raquo; . From there, the farewell of his employees who felt deceived by the about-face on Microsoft's ideals and one billion investments. No longer the research institute that promised to be &ldquo;open&rdquo;, profit growth has become a legal responsibility of the company. His fans in Silicon Valley who had been advocating AGI's oversight in the nonprofit arena lost faith, while investors' Pavlovian response to profit sparked a financial buzz. OpenAI therefore needed to redo its image and DALL-E has become the fun way to regain public favor. DALL-E is silly, trendy, mind-blowing, incredibly viral - and available only to those who respect his method. In the beginning, users were not allowed to sell artwork generated with DALL-E and the underlying model no longer learned anything (i.e. it did not collect the created images as data). But, as a nonprofit, the business model was not sustainable and capable of generating the necessary capital return, and was also subject to change, an evolution already underway.

Since July, DALL-E 2 has gone from paid to free and has removed the ban on the sale of generative art. Updated sessions were released to users, the code of conduct relaxed, resulting in the public release of the generator adding millions more users and generations. While the free version seems antithetical to profit, Tristan Harris, a former Google designer and co-founder of the Center for Humane Technology, explains: "If you don't pay for the product, then you are the product." This model is never for the benefit of the users. The recent changes in DALL-E 2 present some ethical problems: first, the fact of becoming a product, collecting data on the artist's brain and deconstructing human experience as part of the ambitions to develop AGI is never clearly explained. Second, generating work &ldquo;in the style of&rdquo; and then selling that work is ethically dubious and potentially fatal to <a href="https://www.lofficielibiza.com/art/new-young-artists-to-watch-2022-gallery" target="_blank" rel="noopener">artists</a> and their intellectual property. The ability to do &ldquo;in the style of&rdquo; research has dramatic intellectual property implications, and OpenAI's approximately 200 employees lack the tools to address this concern. If an artist owns a creation, does he also own its creative derivatives? If you generate &ldquo;a cat riding a monster in the style of <a href="https://www.lofficielibiza.com/art/jean-michel-basquiat-career-history" target="_blank" rel="noopener">Basquiat</a> &rdquo; . OpenAI does not pay a check to Basquiat's heirs. But what if he did? A layering of commercial rights management could offer control and compensation for that ITSO (In the Style of), but even then there could be unintended consequences. What does it take to determine ITSO? How blurred are the boundaries of "in the style of"? How large is the art style space? How many specific styles can there be? Suppose individual styles have broad boundaries, to make the space of possible styles finite and limited. In that case, this could turn into a rush to set a precedent and reclaim a certain part of a style's space. Artists can have two distinct phases in their careers: first they will need to generate a sufficient level of novelty, get it incorporated into the DALL-E training set, and then apply it to a single style. In the second phase of their careers, they will monetize their style by using the system to generate new materials with their own style vector, or by allowing others to use it to create new art and be compensated passively. If ITSO can be ethically limited, compensation will be required if DALL-E 2 implements a rights management system. Any type of financial approach will require a comparison with the artist, of which at the moment there seems to be little trace. Already now, users are abusing ITSO, leaving artists in the lurch. Artists should have a myriad of creative and commercial options, if ITSO continues as it is, including a usage fee, or rationing a limited number of ITSOs to be obtained through auction. Some artists may require a workflow approval, where the artists approve the artwork and you pay a significant fee. Or an artist can restrict the use of ITSO only to himself and exploit it to scale his production. But this too has its own set of problems. Artists using technology to boost production is not an unfounded dilemma when <a href="https://www.lofficielibiza.com/art/parra-romero" target="_blank" rel="noopener">gallery</a> and art fair schedules require an artist to consistently produce work.

It&#39;s an unfortunate and often undiscussed truth that leads to a particularly bleak moment in art. Such pressure can push artists to use DALL-E 2 to respect the instructions, even though the generator has no subjectivity, or living experience, and uses underlying captions that are not pertinent to the work and, therefore, radically disconnected. Would there be a refusal? A need for transparency? In a lively market, if you keep throwing out what&#39;s selling, is anyone asking the necessary questions? While there may be a distinctive approach to research, the old practice of &quot;finding&quot; and &quot;ruminating&quot; will become increasingly rare. Indeed, the speed of accumulation of a visual vocabulary is not a problem that needs to be solved, but a theme that needs deceleration if a legacy of art is to be guaranteed. If the Maestro envisioned a clock falling off a table, one need only give input to Google, Pinterest, DALL-E 2, or even worse, the open source Stability Diffusion. These technological improvements allow us to bypass the imagination in search of the perfect image, delivered to your desktop at the behest of an algorithm. And this despite AI having no understanding of the human experience, despite the programmer doing their best to replicate the creative singularity (assuming this effort was a priority for the Silicon Valley ethos). It is not argumentative to say that text-to-image generators aim to replace creation with a data-gathering technique, which means that they do not adhere to the ethical and ethereal boundaries that art and humanity have institutionalized. Without ITSO protection, blind contributing to AGI seems misguided and dangerous. In this sense, DALL-E 2 simply amplifies the theme of the current era of commercialized flop, reducing everything to meeting the deadlines required by the galleries and lacking a basic respect for the creative process. Innovation is deeply problematic and, for that reason, it shouldn&#39;t be taken lightly, though it&#39;s incredibly fun. That last point is notable because, as nihilistic as DALL-E 2 might make us, it&#39;s a blast to play.

Silicon Dream: Elon Musk's algorithm that creates images

Tags

Related Articles

Art

DALL-E 2 Presents the Future of Art Through AI

Culture

11 Celebrities Leaving Twitter After the Elon Musk Takeover

Art

How 6 Artists Are Intrepreting the Digital Art Landscape

Recommended posts for you

Fashion

Onitsuka Tiger x Patou: the desirable collaboration of the moment

Beauty

Office hairstyles inspired by Hailey Bieber

Culture

Farewell to Liam Payne: a look back at an emotional ceremony

Culture

Evian and Pharrell Williams unveil their bottle, Fountain of Youth

Fashion

Coquette Aesthetic: The Romantic Trend of Winter 2024

Ibiza

My Island: Sylvie Gianella

Ibiza

Four of the Best Sunday Roasts in Ibiza

Food

Pumpkin Spice Margarita: The Cocktail That Captures the Essence of Autumn