The 2010s bring learned image models into this history alongside continuing developments in creative coding and early blockchain art. GANs and DeepDream change the possibilities of working with images; p5.js extends programming into the browser; artists test ownership and smart contracts as part of their practices. These parallel stories give the chapter its context.
From Le Random’s Generative Art Timeline. Original moment texts by Peter Bauman.
Machine vision becomes a generative resource
The archive follows the 2012 ImageNet result into the introduction of GANs in 2014 and DeepDream in 2015. These entries describe different technical developments: recognizing images, learning to generate samples and visualizing the features a network has learned. Keeping those distinctions visible helps explain the range of artistic experiments gathered under the decade’s AI label.
A block diagram of the AlexNet architecture -- the University of Toronto 'SuperVision' network whose stacked convolution and pooling layers won the 2012 ImageNet challenge. Diagram by Aston Zhang, Zachary C. Lipton, Mu Li and Alexander J. Smola (Dive into Deep Learning), CC BY-SA 4.0, via Wikimedia Commons.
Arguably the biggest development in machine learning in decades, deep convolutional neural networks (CNNs or “ConvNets”) revolutionized machine vision. Their power was deployed for the first time relating to creative expression when the SuperVision algorithm from the University of Toronto used CNNs to win the ImageNet Large-Scale Visual Recognition Challenge. Critically, ConvNets were the driving force behind the decade’s upcoming AI developments This paradigm-shifting…
Figure 2 from Generative Adversarial Nets (2014): generated samples for MNIST, TFD and CIFAR-10, with nearest training examples in the rightmost columns. Ian J. Goodfellow, Jean Pouget-Abadie, Mehdi Mirza, Bing Xu, David Warde-Farley, Sherjil Ozair, Aaron Courville and Yoshua Bengio. Figure 2, Generative Adversarial Nets, 2014, p. 6. Source: arXiv:1406.2661. Image source.
In one of this story’s most relevant moments, Ian Goodfellow and his colleagues at the University of Montreal introduced the world to Generative Adversarial Networks (GANs), revolutionizing—and I don’t use the word lightly—the field of machine learning. GANs consist of two models: a generative model that captures the data distribution, and a discriminative model that estimates the likelihood of a sample being from the…
2015 stands as a monumental year in this story—up there with 1965 (digital art’s public beginnings) and 1995 (net.art)—as the year AI art properly took off. GANs had been introduced a year earlier and served as a primary catalyst along with this seminal moment, the launch of Google’s DeepDream. These events ensured 2015’s place in history with important work by AI legends Gene Kogan,…
A grid of sample nude portraits generated by Robbie Barrat's nude-portrait GAN, published by the artist in the README of his own art-DCGAN repository. Robbie Barrat, sample output from nude-portrait GAN, art-DCGAN, 2018. Courtesy of the artist, via the artist's GitHub repository.
In 2017 Robbie Barrat began using Generative Adversarial Networks to make art, producing his first two significant bodies of work, Landscapes and Nude Portraits, which reached collectors in 2018. He had begun his explorations of AI in 2016, with a Bach-imitating LSTM in July and a rapping neural network in November. His initial attempt with landscapes yielded realistic but ultimately unremarkable results, highlighting the…
AI is one route through this chapter. Another runs through Lauren Lee McCarthy’s p5.js and The Book of Shaders by Patricio González Vivo and Jen Lowe. These projects extend the archive’s history of sharing tools and teaching code. Their presence alongside AI experiments is important: the decade’s generative art includes many technical methods and communities, with different ideas about access and authorship.
The p5.js Web Editor at its 2018 launch, displaying Matthew Kaney’s Tea Time. A later interface for the library initiated in 2013. Screenshot: Processing Foundation. Tea Time: Matthew Kaney. Web Editor development led by Cassie Tarakajian. Image source.
Lauren Lee McCarthy created p5.js roughly twelve years after the initial release of Processing in 2001. P5.js is a JavaScript library that revitalizes the goals of the original Processing project for today’s web environment, where JavaScript—quite different from Java, despite their similar names—has become the dominant Internet-based language. Originating from Processing, designed by Ben Fry and Casey Reas between 2000 and 2001 as a…
Patricio González Vivo and Jen Lowe’s incredibly versatile open-source project, The Book of Shaders, has been published in twelve languages and applies to various environments, including Processing, openFrameworks and Three.js. The online book and toolkit have revolutionized the creative environment since their initial publications by teaching the world about the transformative power of fragment shaders. These short programs offer enhanced control and faster rendering…
Jennifer and Kevin McCoy’s Quantum and Rhea Myers’s Is Art provide distinct early blockchain-art entry points. The former concerns provenance and transferable ownership; the latter uses a contract’s changing art status as part of the work. These artist-led experiments prepare the next chapter’s platforms without making ownership technology a definition of generative art. The 2010s chapter conversation joins Tyler Hobbs, Helena Sarin, Rhea Myers and Gene Kogan.
NFTs Enable Digital Ownership While previous explorations of the blockchain existed prior to Jennifer and Kevin McCoy’s Quantum—many of them can be found previously on this Timeline—none presented a complete package of what NFTs would be than the McCoys. Quantum was also, critically, publicly explained and presented the following afternoon, after it was “minted” on May 2, 2014. This legendary presentation by Kevin McCoy…
This piece from July 23, 2014, under three months after Quantum is another proto-NFT from 2014, yet it derives inspiration from Duchamp, 1960s Conceptualism and net.art. Other early proto-NFTs include NILIcoin and Quantum from the same year, and Mothership from 2013—all found on the Timeline. The work, published as contract code in 2014 and deployed to Ethereum in 2016, that can be toggled between…
This guide connects selected entries from Le Random’s archive of 1,147 moments. Dates follow the individual records. The excerpts, source references and image credits come from those records; follow a moment to read its full account and related interviews.
The archive includes historical precursors and surrounding developments as well as generative artworks. Inclusion does not mean that every object or event was described as “generative art” in its own time.
