artificial intelligence art

{{See also|History of artificial intelligence|Timeline of artificial intelligence}}

File:Maillardet's_automaton_at_the_Franklin_Institute.webm drawing a picture]]

Automated art dates back at least to the automata of ancient Greek civilization, when inventors such as Daedalus and Hero of Alexandria were described as designing machines capable of writing text, generating sounds, and playing music.{{citation | author = Noel Sharkey | title = A programmable robot from 60 AD | date = July 4, 2007 | url = https://www.newscientist.com/blog/technology/2007/07/programmable-robot-from-60ad.html | volume = 2611 | access-date = October 22, 2019 | archive-url = https://web.archive.org/web/20180113090903/https://www.newscientist.com/blog/technology/2007/07/programmable-robot-from-60ad.html | url-status = live | publisher = New Scientist | archive-date = January 13, 2018 }}{{Citation | last = Brett | first = Gerard | title = The Automata in the Byzantine "Throne of Solomon" | date = July 1954 | journal = Speculum | volume = 29 | issue = 3 | pages = 477–487 | postscript = . | doi = 10.2307/2846790 | issn = 0038-7134 | jstor = 2846790 | s2cid = 163031682 }} Creative automatons have flourished throughout history, such as Maillardet's automaton, created around 1800 and capable of creating multiple drawings and poems.{{Cite web | last = kelinich | date = 2014-03-08 | title = Maillardet's Automaton | url = https://www.fi.edu/en/history-resources/automaton | access-date = 2023-08-24 | website = The Franklin Institute | language = en }}

Also in the 19th century, Ada Lovelace, writes that "computing operations" could be used to generate music and poems, now referred to as "The Lovelace Effect," where a computer's behavior is viewed as creative.Natale, S., & Henrickson, L. (2022). The Lovelace Effect: Perceptions of Creativity in Machines. White Rose Research Online. Retrieved September 24, 2024, from https://eprints.whiterose.ac.uk/182906/6/NMS-20-1531.R2_Proof_hi%20%282%29.pdf

Lovelace also discusses a concept known as "The Lovelace Objection," where she argues that a machine has "no pretensions whatever to originate anything."Lovelace, A. (1843). Notes by the translator. Taylor’s Scientific Memoirs, 3, 666-731.

In 1950, with the publication of Alan Turing's paper "Computing Machinery and Intelligence", there was a shift from defining machine intelligence in abstract terms to evaluating whether a machine can mimic human behavior and responses convincingly.{{Cite web |last=Turing |first=Alan |date=October 1950 |title=Computing Machinery and Intelligence |url=https://courses.cs.umbc.edu/471/papers/turing.pdf |access-date=September 16, 2024}} Shortly after, the academic discipline of artificial intelligence was founded at a research workshop at Dartmouth College in 1956.{{Cite book | last = Crevier | first = Daniel | title = AI: The Tumultuous Search for Artificial Intelligence. | publisher = BasicBooks | year = 1993 | isbn = 0-465-02997-3 | location = New York, NY | pages = 109 }} Since its founding, researchers in the field have explored philosophical questions about the nature of the human mind and the consequences of creating artificial beings with human-like intelligence; these issues have previously been explored by myth, fiction, and philosophy since antiquity.{{Cite book | last = Newquist | first = HP | title = The Brain Makers: Genius, Ego, And Greed In The Quest For Machines That Think | publisher = Macmillan/SAMS | year = 1994 | isbn = 978-0-672-30412-5 | location = New York | pages = 45–53 }}

File:Galapagos-icc-2.jpg' Galápagos installation allowed visitors to evolve 3D animated forms]]Since the founding of AI in the 1950s, artists have used artificial intelligence to create artistic works. These works were sometimes referred to as algorithmic art,{{Cite journal | last = Elgammal | first = Ahmed | date = 2019 | title = AI Is Blurring the Definition of Artist | url = http://dx.doi.org/10.1511/2019.107.1.18 | journal = American Scientist | volume = 107 | issue = 1 | pages = 18 | doi = 10.1511/2019.107.1.18 | s2cid = 125379532 | issn = 0003-0996 }} computer art, digital art, or new media art.{{Cite journal | last = Greenfield | first = Gary | date = 2015-04-03 | title = When the machine made art: the troubled history of computer art, by Grant D. Taylor | url = http://www.tandfonline.com/doi/full/10.1080/17513472.2015.1009865 | journal = Journal of Mathematics and the Arts | language = en | volume = 9 | issue = 1–2 | pages = 44–47 | doi = 10.1080/17513472.2015.1009865 | s2cid = 118762731 | issn = 1751-3472 }}

One of the first significant AI art systems is AARON, developed by Harold Cohen beginning in the late 1960s at the University of California at San Diego.{{Cite book | last = McCorduck | first = Pamela | title = AARONS's Code: Meta-Art. Artificial Intelligence, and the Work of Harold Cohen | publisher = W. H. Freeman and Company | year = 1991 | isbn = 0-7167-2173-2 | location = New York | pages = 210 | language = English }} AARON uses a symbolic rule-based approach to generate technical images in the era of GOFAI programming, and it was developed by Cohen with the goal of being able to code the act of drawing.{{Cite book | last1 = Poltronieri | first1 = Fabrizio Augusto | last2 = Hänska | first2 = Max | title = Proceedings of the 9th International Conference on Digital and Interactive Arts | chapter = Technical Images and Visual Art in the Era of Artificial Intelligence | date = 2019-10-23 | chapter-url = https://dl.acm.org/doi/10.1145/3359852.3359865 | language = en | location = Braga Portugal | publisher = ACM | pages = 1–8 | doi = 10.1145/3359852.3359865 | isbn = 978-1-4503-7250-3 | s2cid = 208109113 }} AARON was exhibited in 1972 at the Los Angeles County Museum of Art.{{Cite web | date = May 9, 2016 | title = HAROLD COHEN (1928–2016) | url = https://www.artforum.com/news/harold-cohen-1928-2016-59932 | access-date = 2023-09-19 | website = Art Forum | language = en-US }} From 1973 to 1975, Cohen refined AARON during a residency at the Artificial Intelligence Laboratory at Stanford University.{{Cite news |last=Diehl |first=Travis |date=2024-02-15 |title=A.I. Art That's More Than a Gimmick? Meet AARON |url=https://www.nytimes.com/2024/02/15/arts/design/aaron-ai-whitney.html |access-date=2024-06-01 |work=The New York Times |language=en-US |issn=0362-4331}} In 2024, the Whitney Museum of American Art exhibited AI art from throughout Cohen's career, including re-created versions of his early robotic drawing machines.

Karl Sims has exhibited art created with artificial life since the 1980s. He received an M.S. in computer graphics from the MIT Media Lab in 1987 and was artist-in-residence from 1990 to 1996 at the supercomputer manufacturer and artificial intelligence company Thinking Machines.{{Cite web |date=2017-08-20 |title=Karl Sims - ACM SIGGRAPH HISTORY ARCHIVES |url=https://history.siggraph.org/person/karl-sims/ |access-date=2024-06-09 |website=history.siggraph.org |language=en-US}}{{Cite web |title=Karl Sims {{!}} CSAIL Alliances |url=https://cap.csail.mit.edu/engage/spotlights/karl-sims |access-date=2024-06-09 |website=cap.csail.mit.edu |language=en}}{{Cite web |title=Karl Sims |url=https://www.macfound.org/fellows/class-of-1998/karl-sims |access-date=2024-06-09 |website=www.macfound.org |language=en}} In both 1991 and 1992, Sims won the Golden Nica award at Prix Ars Electronica for his videos using artificial evolution.{{Cite web |title = Golden Nicas |url = https://ars.electronica.art/center/en/golden-nicas/ |access-date = 2023-02-26 |website = Ars Electronica Center |language = en-US |archive-date = 26 February 2023 |archive-url = https://web.archive.org/web/20230226175530/https://ars.electronica.art/center/en/golden-nicas/ |url-status = dead}}{{Cite web | title = Panspermia by Karl Sims, 1990 | url = http://www.karlsims.com/panspermia.html | access-date = 2023-02-26 | website = www.karlsims.com }}{{Cite web | title = Liquid Selves by Karl Sims, 1992 | url = http://www.karlsims.com/liquid-selves.html | access-date = 2023-02-26 | website = www.karlsims.com }} In 1997, Sims created the interactive artificial evolution installation Galápagos for the NTT InterCommunication Center in Tokyo.{{Cite web |title=ICC {{!}} "Galápagos" - Karl SIMS (1997) |url=https://www.ntticc.or.jp/en/archive/works/galapagos/ |access-date=2024-06-14 |website=NTT InterCommunication Center [ICC] |language=en}} Sims received an Emmy Award in 2019 for outstanding achievement in engineering development.{{Cite web |title=- Winners |url=https://www.emmys.com/awards/engineering-emmys/winners |access-date=2022-06-26 |website=Television Academy |language=en}} File:Still from Eric Millikin's "The Dance of the Nain Rouge," with subtitles.jpg's The Dance of the Nain Rouge, with subtitles]]Eric Millikin has been creating animated films using artificial intelligence since the early 1980s, posting the art on the internet using CompuServe.{{cite news |last1=Baetens |first1=Melody |date=October 25, 2023 |title=Things to do this Halloween weekend in Metro Detroit |url=https://www.detroitnews.com/story/entertainment/2023/10/25/things-to-do-this-halloween-weekend-in-metro-detroit/71313694007/ |work=The Detroit News}}{{cite web |last1=Angelo |first1=Delos Trinos |date=26 October 2023 |title=10 Greatest Innovations In Comics History |url=https://www.cbr.com/comics-history-greatest-creations/ |access-date=1 November 2023 |website=Comic Book Resources}} Millikin also released the AI film The Dance of the Nain Rouge in 2023, about the Detroit folklore legend of the Nain Rouge. The film is described as "an experimental decolonial Detroit demonology deepfake dream dance documentary" and has won awards at several film festivals.{{cite web |last1=Ringler |first1=Chris |date=18 October 2022 |title=THE DANCE OF THE NAIN ROUGE |url=https://spookychris.com/flint-monster-society/flint-short-film-freakout-films/ |access-date=1 November 2023}}{{Cite web |title=PISA ROBOT FILM FESTIVAL 3 - I vincitori - CinemaItaliano.info |url=https://www.cinemaitaliano.info/news/79524/pisa-robot-film-festival-3-i-vincitori.html |access-date=2024-06-03 |website=www.cinemaitaliano.info}}{{Cite web |last= |date=2024-02-01 |title=Awards of December 2023 – January 2024 |url=https://www.assurdofilmfestival.com/news/december-2023-january-2024/ |access-date=2024-03-31 |website=Absurd Film Festival |language=it-IT}} File:Electricsheep-0-1000.jpg]]In 1999, Scott Draves and a team of several engineers created and released Electric Sheep as a free software screensaver.{{Cite book |last=Draves |first=Scott |title=Applications of Evolutionary Computing |date=2005 |publisher=Springer |isbn=978-3-540-32003-6 |editor-last=Rothlauf |editor-first=Franz |series=Lecture Notes in Computer Science |volume=3449 |location=Berlin, Heidelberg |pages=458–467 |language=en |chapter=The Electric Sheep Screen-Saver: A Case Study in Aesthetic Evolution |doi=10.1007/978-3-540-32003-6_46 |editor2-last=Branke |editor2-first=Jürgen |editor3-last=Cagnoni |editor3-first=Stefano |editor4-last=Corne |editor4-first=David Wolfe |editor5-last=Drechsler |editor5-first=Rolf |editor6-last=Jin |editor6-first=Yaochu |editor7-last=Machado |editor7-first=Penousal |editor8-last=Marchiori |editor8-first=Elena |editor9-last=Romero |editor9-first=Juan |chapter-url=https://link.springer.com/chapter/10.1007/978-3-540-32003-6_46 |s2cid=14256872}} Electric Sheep is a volunteer computing project for animating and evolving fractal flames, which are distributed to networked computers which display them as a screensaver. The screensaver used AI to create an infinite animation by learning from its audience. In 2001, Draves won the Fundacion Telefónica Life 4.0 prize for Electric Sheep.{{Cite web |title=Entrevista Scott Draves - Primer Premio Ex-Aequo VIDA 4.0 |website=YouTube |date=17 July 2012 |url=https://www.youtube.com/watch?v=wybvI279EQ4 |access-date=2023-02-26}}{{Unreliable source?|date=April 2025}}

In 2014, Stephanie Dinkins began working on Conversations with Bina48.{{Cite web |date=7 November 2017 |title=Robots, Race, and Algorithms: Stephanie Dinkins at Recess Assembly |url=http://magazine.art21.org/2017/11/07/robots-race-and-algorithms-stephanie-dinkins-at-recess-assembly/ |access-date=2020-02-25 |website=Art21 Magazine}} For the series, Dinkins recorded her conversations with BINA48, a social robot that resembles a middle-aged black woman.{{Cite web |last=Small |first=Zachary |date=2017-04-07 |title=Future Perfect: Flux Factory's Intersectional Approach to Technology |url=https://www.artnews.com/art-in-america/features/future-perfect-a-call-for-intersectional-technology-at-flux-factory-58475/ |access-date=2020-05-04 |website=ARTnews.com |language=en-US}}{{Cite web |last=Dunn |first=Anna |date=July 11, 2018 |title=Multiply, Identify, Her |url=https://brooklynrail.org/2018/07/artseen/Multiply-Identify-Her |website=The Brooklyn Rail}} In 2019, Dinkins won the Creative Capital award for her creation of an evolving artificial intelligence based on the "interests and culture(s) of people of color."{{Cite web |title=Not the Only One |url=https://creative-capital.org/projects/not-the-only-one/ |access-date=2023-02-26 |website=Creative Capital |language=en}}

In 2015, Sougwen Chung began Mimicry (Drawing Operations Unit: Generation 1), an ongoing collaboration between the artist and a robotic arm.{{Cite web |title=Drawing Operations (2015) – Sougwen Chung (愫君) |url=https://sougwen.com/project/drawing-operations |access-date=2025-02-25 |language=en-US}} In 2019, Chung won the Lumen Prize for her continued performances with a robotic arm that uses AI to attempt to draw in a manner similar to Chung.{{Cite web |title=Sougwen Chung |url=https://www.lumenprize.com/2019-winners/sougwen-chung |access-date=2023-02-26 |website=The Lumen Prize |language=en-GB}} File:Edmond de Belamy.png, created with a generative adversarial network in 2018]]In 2018, an auction sale of artificial intelligence art was held at Christie's in New York where the AI artwork Edmond de Belamy sold for {{Currency|432,500|USD}}, which was almost 45 times higher than its estimate of {{Currency|7,000|USD|linked=no}}–10,000. The artwork was created by Obvious, a Paris-based collective.{{cite web |date=2018-12-12 |title=Is artificial intelligence set to become art's next medium? |url=https://www.christies.com/features/A-collaboration-between-two-artists-one-human-one-a-machine-9332-1.aspx |access-date=2019-05-21 |website=Christie's}}{{Cite news |last=Cohn |first=Gabe |date=2018-10-25 |title=AI Art at Christie's Sells for $432,500 |url=https://www.nytimes.com/2018/10/25/arts/design/ai-art-sold-christies.html |url-status=live |archive-url=https://web.archive.org/web/20190505102713/https://www.nytimes.com/2018/10/25/arts/design/ai-art-sold-christies.html |archive-date=5 May 2019 |access-date=2024-05-26 |newspaper=The New York Times |language=en-US |issn=0362-4331}}{{Cite web |last=Turnbull |first=Amanda |date=2020-01-06 |title=The price of AI art: Has the bubble burst? |url=http://theconversation.com/the-price-of-ai-art-has-the-bubble-burst-128698 |url-status=live |archive-url=https://web.archive.org/web/20240526121500/http://theconversation.com/the-price-of-ai-art-has-the-bubble-burst-128698 |archive-date=26 May 2024 |access-date=2024-05-26 |website=The Conversation |language=en-US}}

In 2024, Japanese film generAIdoscope was released. The film was co-directed by Hirotaka Adachi, Takeshi Sone, and Hiroki Yamaguchi. All video, audio, and music in the film were created with artificial intelligence.{{Cite web |last=Cayanan |first=Joanna |date=2024-07-13 |title=Novelist Otsuichi Co-Directs generAIdoscope, Omnibus Film Produced Entirely With Generative AI |url=https://www.animenewsnetwork.com/news/2024-07-13/novelist-otsuichi-co-directs-generaidoscope-omnibus-film-produced-entirely-with-generative-ai/.213069 |access-date=2025-03-04 |website=Anime News Network |language=en}}

In 2025, Japanese anime television series Twins Hinahima was released. The anime was produced and animated with AI assistance during the process of cutting and conversion of photographs into anime illustrations and later retouched by art staff. Most of the remaining parts such as characters and logos were hand-drawn with various software.{{Cite web |last=Hodgkins |first=Crystalyn |date=2025-02-28 |title=Frontier Works, KaKa Creation's Twins Hinahima AI Anime Reveals March 29 TV Debut |url=https://www.animenewsnetwork.com/news/2025-02-28/frontier-works-kaka-creation-twins-hinahima-ai-anime-reveals-march-29-tv-debut/.221769 |access-date=2025-03-04 |website=Anime News Network |language=en}}{{Cite web |title=サポーティブAIとは - アニメ「ツインズひなひま」公式サイト |trans-title=What's Supportive AI? - Twins Hinahima Anime Official Website |url=https://anime-hinahima.com/supportive-ai/ |access-date=2025-03-04 |website=anime-hinahima.com |language=ja}}

Deep learning, characterized by its multi-layer structure that attempts to mimic the human brain, first came about in the 2010s and causing a significant shift in the world of AI art.{{Cite web |date=2024-06-17 |title=What Is Deep Learning? {{!}} IBM |url=https://www.ibm.com/topics/deep-learning |access-date=2024-11-13 |website=www.ibm.com |language=en}} During the deep learning era, there are mainly these types of designs for generative art: autoregressive models, diffusion models, GANs, normalizing flows.

In 2014, Ian Goodfellow and colleagues at Université de Montréal developed the generative adversarial network (GAN), a type of deep neural network capable of learning to mimic the statistical distribution of input data such as images. The GAN uses a "generator" to create new images and a "discriminator" to decide which created images are considered successful.{{cite conference | last1 = Goodfellow | first1 = Ian | last2 = Pouget-Abadie | first2 = Jean | last3 = Mirza | first3 = Mehdi | last4 = Xu | first4 = Bing | last5 = Warde-Farley | first5 = David | last6 = Ozair | first6 = Sherjil | last7 = Courville | first7 = Aaron | last8 = Bengio | first8 = Yoshua | year = 2014 | title = Generative Adversarial Nets | url = https://papers.nips.cc/paper/5423-generative-adversarial-nets.pdf | conference = Proceedings of the International Conference on Neural Information Processing Systems (NIPS 2014) | pages = 2672–2680 }} Unlike previous algorithmic art that followed hand-coded rules, generative adversarial networks could learn a specific aesthetic by analyzing a dataset of example images.

In 2015, a team at Google released DeepDream, a program that uses a convolutional neural network to find and enhance patterns in images via algorithmic pareidolia.{{cite web | last1 = Mordvintsev | first1 = Alexander | last2 = Olah | first2 = Christopher | last3 = Tyka | first3 = Mike | year = 2015 | title = DeepDream - a code example for visualizing Neural Networks | url = http://googleresearch.blogspot.co.uk/2015/07/deepdream-code-example-for-visualizing.html | archive-url = https://web.archive.org/web/20150708233542/http://googleresearch.blogspot.co.uk/2015/07/deepdream-code-example-for-visualizing.html | archive-date = 2015-07-08 | publisher = Google Research }}{{cite web | last1 = Mordvintsev | first1 = Alexander | last2 = Olah | first2 = Christopher | last3 = Tyka | first3 = Mike | year = 2015 | title = Inceptionism: Going Deeper into Neural Networks | url = http://googleresearch.blogspot.co.uk/2015/06/inceptionism-going-deeper-into-neural.html | archive-url = https://web.archive.org/web/20150703064823/http://googleresearch.blogspot.co.uk/2015/06/inceptionism-going-deeper-into-neural.html | archive-date = 2015-07-03 | publisher = Google Research }}{{cite conference | last1 = Szegedy | first1 = Christian | last2 = Liu | first2 = Wei | last3 = Jia | first3 = Yangqing | last4 = Sermanet | first4 = Pierre | last5 = Reed | first5 = Scott E. | last6 = Anguelov | first6 = Dragomir | last7 = Erhan | first7 = Dumitru | last8 = Vanhoucke | first8 = Vincent | last9 = Rabinovich | first9 = Andrew | arxiv = 1409.4842 | contribution = Going deeper with convolutions | doi = 10.1109/CVPR.2015.7298594 | pages = 1–9 | publisher = IEEE Computer Society | title = IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2015, Boston, MA, USA, June 7–12, 2015 | year = 2015 | isbn = 978-1-4673-6964-0 }} The process creates deliberately over-processed images with a dream-like appearance reminiscent of a psychedelic experience.{{cite web | first1 = Alexander | last1 = Mordvintsev | first2 = Christopher | last2 = Olah | first3 = Mike | last3 = Tyka | year = 2015 | archive-url = https://web.archive.org/web/20150708233542/http://googleresearch.blogspot.co.uk/2015/07/deepdream-code-example-for-visualizing.html | archive-date = 2015-07-08 | url = http://googleresearch.blogspot.co.uk/2015/07/deepdream-code-example-for-visualizing.html | title = DeepDream - a code example for visualizing Neural Networks | publisher = Google Research }} Later, in 2017, a conditional GAN learned to generate 1000 image classes of ImageNet, a large visual database designed for use in visual object recognition software research.{{Cite web |last=Reynolds |first=Matt |date=7 April 2017 |title=New computer vision challenge wants to teach robots to see in 3D |url=https://www.newscientist.com/article/2127131-new-computer-vision-challenge-wants-to-teach-robots-to-see-in-3d/ |access-date=2024-11-15 |website=New Scientist |language=en-US}}{{Cite web |last=Markoff |first=John |date=19 November 2012 |title=Seeking a Better Way to Find Web Images |website=The New York Times |url=https://www.nytimes.com/2012/11/20/science/for-web-images-creating-new-technology-to-seek-and-find.html?smid=url-share}} By conditioning the GAN on both random noise and a specific class label, this approach enhanced the quality of image synthesis for class-conditional models.{{Cite journal |last1=Odena |first1=Augustus |last2=Olah |first2=Christopher |last3=Shlens |first3=Jonathon |date=2017-07-17 |title=Conditional Image Synthesis with Auxiliary Classifier GANs |url=https://proceedings.mlr.press/v70/odena17a.html |journal=International Conference on Machine Learning |language=en |publisher=PMLR |pages=2642–2651 |arxiv=1610.09585}}

Autoregressive models were used for image generation, such as PixelRNN (2016), which autoregressively generates one pixel after another with a recurrent neural network.{{Cite journal |last1=Oord |first1=Aäron van den |last2=Kalchbrenner |first2=Nal |last3=Kavukcuoglu |first3=Koray |date=2016-06-11 |title=Pixel Recurrent Neural Networks |url=https://proceedings.mlr.press/v48/oord16.html |journal=Proceedings of the 33rd International Conference on Machine Learning |language=en |publisher=PMLR |pages=1747–1756}} Immediately after the Transformer architecture was proposed in Attention Is All You Need (2018), it was used for autoregressive generation of images, but without text conditioning.{{Cite journal |last1=Parmar |first1=Niki |last2=Vaswani |first2=Ashish |last3=Uszkoreit |first3=Jakob |last4=Kaiser |first4=Lukasz |last5=Shazeer |first5=Noam |last6=Ku |first6=Alexander |last7=Tran |first7=Dustin |date=2018-07-03 |title=Image Transformer |url=https://proceedings.mlr.press/v80/parmar18a.html |journal=Proceedings of the 35th International Conference on Machine Learning |language=en |publisher=PMLR |pages=4055–4064}}

The website Artbreeder, launched in 2018, uses the models StyleGAN and BigGAN{{cite web | url = https://www.artbreeder.com/about | title = About | access-date = March 3, 2021 | last = Simon | first = Joel | archive-url = https://web.archive.org/web/20210302075357/https://www.artbreeder.com/about | archive-date = March 2, 2021 | url-status = live }}{{Cite book | first1 = Binto | last1 = George | first2 = Gail | last2 = Carmichael | title = Artificial Intelligence Simplified: Understanding Basic Concepts -- the Second Edition | pages = 7–25 | url = https://books.google.com/books?id=duQaEAAAQBAJ | editor-first = Susan | editor-last = Mathai | isbn = 9781944708047 | date = 2021 | publisher = CSTrends LLP }} to allow users to generate and modify images such as faces, landscapes, and paintings.{{cite web | url = https://www.digitalartsonline.co.uk/news/creative-software/will-this-creepy-ai-platform-put-artists-out-of-job/ | title = Will this creepy AI platform put artists out of a job? | access-date = March 3, 2021 | last = Lee | first = Giacomo | date = July 21, 2020 | archive-url = https://web.archive.org/web/20201222214934/https://www.digitalartsonline.co.uk/news/creative-software/will-this-creepy-ai-platform-put-artists-out-of-job/ | archive-date = December 22, 2020 | website = Digital Arts Online | url-status = live }}

In the 2020s, text-to-image models, which generate images based on prompts, became widely used, marking yet another shift in the creation of AI generated artworks.

{{Multiple image

| direction = vertical

| total_width = 300

| image1 = Scenic Valley in the Afternoon Artistic (VQGAN+CLIP).jpg

| alt1 =

| caption1 = Example of an image made with VQGAN-CLIP (NightCafe Studio)

| image2 = Sunset Valley (FLUX 1.1 Pro Raw).webp

| caption2 = Example of an image made with Flux 1.1 Pro in Raw mode; this mode is designed to generate photorealistic images

}}

In 2021, using the influential large language generative pre-trained transformer models that are used in GPT-2 and GPT-3, OpenAI released a series of images created with the text-to-image AI model DALL-E 1.{{cite arXiv |last1=Ramesh |first1=Aditya |last2=Pavlov |first2=Mikhail |last3=Goh |first3=Gabriel |last4=Gray |first4=Scott |last5=Voss |first5=Chelsea |last6=Radford |first6=Alec |last7=Chen |first7=Mark |last8=Sutskever |first8=Ilya |eprint=2102.12092 |title=Zero-Shot Text-to-Image Generation |class=cs.LG |date=24 February 2021}} It was an autoregressive generative model with essentially the same architecture as GPT-3. Along with this, later in 2021, EleutherAI released the open source VQGAN-CLIP{{cite web | last1 = Burgess | first1 = Phillip | title = Generating AI "Art" with VQGAN+CLIP | url = https://learn.adafruit.com/generating-ai-art-with-vqgan-clip | access-date = July 20, 2022 | website = Adafruit }} based on OpenAI's CLIP model.{{cite arXiv |last1=Radford |first1=Alec |last2=Kim |first2=Jong Wook |last3=Hallacy |first3=Chris |last4=Ramesh |first4=Aditya |last5=Goh |first5=Gabriel |last6=Agarwal |first6=Sandhini |last7=Sastry |first7=Girish |last8=Askell |first8=Amanda |last9=Mishkin |first9=Pamela |last10=Clark |first10=Jack |last11=Krueger |first11=Gretchen |last12=Sutskever |first12=Ilya |title=Learning Transferable Visual Models From Natural Language Supervision |year=2021 |class=cs.CV |eprint=2103.00020}} Diffusion models, generative models used to create synthetic data based on existing data,{{Cite web |date=2024-04-04 |title=What Are Diffusion Models? |url=https://www.coursera.org/articles/diffusion-models |access-date=2024-11-13 |website=Coursera |language=en}} were first proposed in 2015,{{Cite journal |last1=Sohl-Dickstein |first1=Jascha |last2=Weiss |first2=Eric |last3=Maheswaranathan |first3=Niru |last4=Ganguli |first4=Surya |date=2015-06-01 |title=Deep Unsupervised Learning using Nonequilibrium Thermodynamics |url=http://proceedings.mlr.press/v37/sohl-dickstein15.pdf |journal=Proceedings of the 32nd International Conference on Machine Learning |language=en |publisher=PMLR |volume=37 |pages=2256–2265|arxiv=1503.03585 }} but they only became better than GANs in early 2021.{{Cite journal |last1=Dhariwal |first1=Prafulla |last2=Nichol |first2=Alexander |date=2021 |title=Diffusion Models Beat GANs on Image Synthesis |url=https://proceedings.neurips.cc/paper/2021/hash/49ad23d1ec9fa4bd8d77d02681df5cfa-Abstract.html |journal=Advances in Neural Information Processing Systems |publisher=Curran Associates, Inc. |volume=34 |pages=8780–8794|arxiv=2105.05233 }} Latent diffusion model was published in December 2021 and became the basis for the later Stable Diffusion (August 2022).{{Citation |last1=Rombach |first1=Robin |title=High-Resolution Image Synthesis with Latent Diffusion Models |date=2021-12-20 |arxiv=2112.10752 |last2=Blattmann |first2=Andreas |last3=Lorenz |first3=Dominik |last4=Esser |first4=Patrick |last5=Ommer |first5=Björn}}

In 2022, Midjourney{{cite news |last1=Rose |first1=Janus |title=Inside Midjourney, The Generative Art AI That Rivals DALL-E |url=https://www.vice.com/en/article/wxn5wn/inside-midjourney-the-generative-art-ai-that-rivals-dall-e |publisher=Vice |date=July 18, 2022}} was released, followed by Google Brain's Imagen and Parti, which were announced in May 2022, Microsoft's NUWA-Infinity,{{Cite web | title = NUWA-Infinity | url = https://nuwa-infinity.microsoft.com/#/ | access-date = 2022-08-10 | website = nuwa-infinity.microsoft.com }} and the source-available Stable Diffusion, which was released in August 2022.{{Cite web | title = Diffuse The Rest - a Hugging Face Space by huggingface | url = https://huggingface.co/spaces/huggingface/diffuse-the-rest | access-date = 2022-09-05 | website = huggingface.co | archive-date = 2022-09-05 | archive-url = https://web.archive.org/web/20220905141431/https://huggingface.co/spaces/huggingface/diffuse-the-rest | url-status = live }}{{cite web | author-last = Heikkilä | author-first = Melissa | date = 16 September 2022 | title = This artist is dominating AI-generated art. And he's not happy about it. | url = https://www.technologyreview.com/2022/09/16/1059598/this-artist-is-dominating-ai-generated-art-and-hes-not-happy-about-it/ | access-date = 2 October 2022 | work = MIT Technology Review }}{{cite web | date = 15 September 2022 | title = Stable Diffusion | url = https://github.com/CompVis/stable-diffusion | access-date = 15 September 2022 | publisher = CompVis - Machine Vision and Learning LMU Munich }} DALL-E{{Nbsp}}2, a successor to DALL-E, was beta-tested and released (with the further successor DALL-E{{Nbsp}}3 being released in 2023). Stability AI has a Stable Diffusion web interface called DreamStudio,{{cite news | date = 18 October 2022 | title = Stable Diffusion creator Stability AI accelerates open-source AI, raises $101M | work = VentureBeat | url = https://venturebeat.com/ai/stable-diffusion-creator-stability-ai-raises-101m-funding-to-accelerate-open-source-ai/ | access-date = 10 November 2022 }} plugins for Krita, Photoshop, Blender, and GIMP,{{cite web | last1 = Choudhary | first1 = Lokesh | date = 23 September 2022 | title = These new innovations are being built on top of Stable Diffusion | url = https://analyticsindiamag.com/these-new-innovations-are-being-built-on-top-of-stable-diffusion/ | access-date = 9 November 2022 | website = Analytics India Magazine }} and the Automatic1111 web-based open source user interface.{{cite news | author1 = Dave James | date = 27 October 2022 | title = I thrashed the RTX 4090 for 8 hours straight training Stable Diffusion to paint like my uncle Hermann | language = en | work = PC Gamer | url = https://www.pcgamer.com/nvidia-rtx-4090-stable-diffusion-training-aharon-kahana/ | access-date = 9 November 2022 }}{{cite web | last1 = Lewis | first1 = Nick | title = How to Run Stable Diffusion Locally With a GUI on Windows | url = https://www.howtogeek.com/832491/how-to-run-stable-diffusion-locally-with-a-gui-on-windows/ | access-date = 9 November 2022 | website = How-To Geek | date = 16 September 2022 }}{{cite news | last1 = Edwards | first1 = Benj | date = 4 October 2022 | title = Begone, polygons: 1993's Virtua Fighter gets smoothed out by AI | language = en-us | work = Ars Technica | url = https://arstechnica.com/gaming/2022/10/begone-polygons-1993s-virtua-fighter-gets-smoothed-out-by-ai/ | access-date = 9 November 2022 }} Stable Diffusion's main pre-trained model is shared on the Hugging Face Hub.{{cite news | last1 = Mehta | first1 = Sourabh | date = 17 September 2022 | title = How to Generate an Image from Text using Stable Diffusion in Python | work = Analytics India Magazine | url = https://analyticsindiamag.com/how-to-generate-an-image-from-text-using-stable-diffusion-on-python/ | access-date = 16 November 2022 }}

Ideogram was released in August 2023, this model is known for its ability to generate legible text.{{cite web |title=Announcing Ideogram AI |url=https://ideogram.ai/launch |access-date=June 13, 2024 |website=Ideogram}}{{Cite news |last=Metz |first=Rachel |date=2023-10-03 |title=Ideogram Produces Text in AI Images That You Can Actually Read |url=https://www.bloomberg.com/news/articles/2023-10-03/ideogram-produces-text-in-ai-images-that-you-can-actually-read |access-date=2024-11-18 |work=Bloomberg News |language=en}}

In 2024, Flux was released. This model can generate realistic images and was integrated into Grok, the chatbot used on X (formerly Twitter), and Le Chat, the chatbot of Mistral AI.{{Cite web |title=Flux.1 – ein deutscher KI-Bildgenerator dreht mit Grok frei |url=https://www.handelsblatt.com/technik/ki/xai-kooperation-flux1-deutscher-ki-bildgenerator-dreht-mit-grok-frei/100059178.html |access-date=2024-11-17 |website=Handelsblatt |language=de}}{{Cite web |last=Zeff |first=Maxwell |date=2024-08-14 |title=Meet Black Forest Labs, the startup powering Elon Musk's unhinged AI image generator |url=https://techcrunch.com/2024/08/14/meet-black-forest-labs-the-startup-powering-elon-musks-unhinged-ai-image-generator/ |access-date=2024-11-17 |website=TechCrunch |language=en-US}}{{Cite web |last=Franzen |first=Carl |date=2024-11-18 |title=Mistral unleashes Pixtral Large and upgrades Le Chat into full-on ChatGPT competitor |url=https://venturebeat.com/ai/mistral-unleashes-pixtral-large-and-upgrades-le-chat-into-full-on-chatgpt-competitor/ |access-date=2024-12-11 |website=VentureBeat |language=en-US}} Flux was developed by Black Forest Labs, founded by the researchers behind Stable Diffusion.{{Cite web |last=Growcoot |first=Matt |date=2024-08-05 |title=AI Image Generator Made by Stable Diffusion Inventors on Par With Midjourney and DALL-E |url=https://petapixel.com/2024/08/05/ai-image-generator-made-by-stable-diffusion-inventors-on-par-with-midjourney-and-dall-e-flux1-black-forest-labs/ |access-date=2024-11-17 |website=PetaPixel |language=en}} Grok later switched to its own text-to-image model Aurora in December of the same year.{{Cite web |last=Davis |first=Wes |date=2024-12-07 |title=X gives Grok a new photorealistic AI image generator |url=https://www.theverge.com/2024/12/7/24315644/grok-x-aurora-ai-image-generator-xai |access-date=2024-12-10 |website=The Verge |language=en}} Several companies, along with their products, have also developed an AI model integrated with an image editing service. Adobe has released and integrated the AI model Firefly into Premiere Pro, Photoshop, and Illustrator.{{Cite web |last=Clark |first=Pam |date=October 14, 2024 |title=Photoshop delivers powerful innovation for Image Editing, Ideation, 3D Design, and more |url=https://blog.adobe.com/en/publish/2024/10/14/photoshop-delivers-powerful-innovation-for-image-editing-ideation-3d-design-more |access-date=2025-02-08 |website=Adobe Blog}}{{Cite web |last=Chedraoui |first=Katelyn |date=October 19, 2024 |title=Every New Feature Adobe Announced in Photoshop, Premiere Pro and More |url=https://www.cnet.com/tech/services-and-software/every-new-feature-adobe-announced-in-photoshop-premiere-pro-and-more/ |access-date=2025-02-08 |website=CNET |language=en}} Microsoft has also publicly announced AI image-generator features for Microsoft Paint.{{Cite web |last=Fajar |first=Aditya |date=2023-08-28 |title=Microsoft Paint will use AI in Windows update 11 |url=https://gizmologi.id/en/application/microsoft-paint-gunakan-ai/ |access-date=2025-02-08 |website=gizmologi.id |language=en}} Along with this, some examples of text-to-video models of the mid-2020s are Runway's Gen-2, Google's VideoPoet, and OpenAI's Sora, which was released in December 2024.{{Cite web |date=2024-02-15 |title=OpenAI teases 'Sora,' its new text-to-video AI model |url=https://www.nbcnews.com/tech/tech-news/openai-sora-video-artificial-intelligence-unveiled-rcna139065 |access-date=2024-10-28 |website=NBC News |language=en}}{{Cite web |title=Sora |url=https://sora.com/ |access-date=2024-12-27 |website=Sora |language=en}}

File:Generic SDXL ComfyUI Nodes (2024) (cropped).png for Stable Diffusion XL, people can adjust variables (such as CFG, seed, and sampler) needed to generate image]]

There are many tools available to the artist when working with diffusion models. They can define both positive and negative prompts, but they are also afforded a choice in using (or omitting the use of) VAEs, LoRAs, hypernetworks, IP-adapter, and embedding/textual inversions. Artists can tweak settings like guidance scale (which balances creativity and accuracy), seed (to control randomness), and upscalers (to enhance image resolution), among others. Additional influence can be exerted during pre-inference by means of noise manipulation, while traditional post-processing techniques are frequently used post-inference. People can also train their own models.

In addition, procedural "rule-based" generation of images using mathematical patterns, algorithms that simulate brush strokes and other painted effects, and deep learning algorithms such as generative adversarial networks (GANs) and transformers have been developed. Several companies have released apps and websites that allow one to forego all the options mentioned entirely while solely focusing on the positive prompt. There also exist programs which transform photos into art-like images in the style of well-known sets of paintings.{{cite news | title = A.I. photo filters use neural networks to make photos look like Picassos | url = https://www.digitaltrends.com/mobile/best-ai-based-photo-apps/ | access-date = 9 November 2022 | work = Digital Trends | date = 18 November 2019 | language = en }}{{cite news | last1 = Biersdorfer | first1 = J. D. | title = From Camera Roll to Canvas: Make Art From Your Photos | url = https://www.nytimes.com/2019/12/04/technology/personaltech/turn-photos-into-paintings.html | access-date = 9 November 2022 | work = The New York Times | date = 4 December 2019 }}

There are many options, ranging from simple consumer-facing mobile apps to Jupyter notebooks and web UIs that require powerful GPUs to run effectively.{{cite web | url = https://pharmapsychotic.com/tools.html | title = Tools and Resources for AI Art | first = Pharma | last = Psychotic | archive-url = https://archive.today/20220604120005/https://pharmapsychotic.com/tools.html | archive-date = 2022-06-04 | access-date = 2022-06-26 }} Additional functionalities include "textual inversion," which refers to enabling the use of user-provided concepts (like an object or a style) learned from a few images. Novel art can then be generated from the associated word(s) (the text that has been assigned to the learned, often abstract, concept){{Cite arXiv|last1=Gal |first1=Rinon |last2=Alaluf |first2=Yuval |last3=Atzmon |first3=Yuval |last4=Patashnik |first4=Or |last5=Bermano |first5=Amit H. |last6=Chechik |first6=Gal |last7=Cohen-Or |first7=Daniel |title=An Image is Worth One Word: Personalizing Text-to-Image Generation using Textual Inversion |date=2 August 2022|class=cs.CV |eprint=2208.01618}}{{cite web | title = Textual Inversion · AUTOMATIC1111/stable-diffusion-webui Wiki | url = https://github.com/AUTOMATIC1111/stable-diffusion-webui/wiki/Textual-Inversion | website = GitHub | access-date = 9 November 2022 | language = en }} and model extensions or fine-tuning (such as DreamBooth).

AI has the potential for a societal transformation, which may include enabling the expansion of noncommercial niche genres (such as cyberpunk derivatives like solarpunk) by amateurs, novel entertainment, fast prototyping,{{cite news | last1 = Elgan | first1 = Mike | date = 1 November 2022 | title = How 'synthetic media' will transform business forever | language = en | work = Computerworld | url = https://www.computerworld.com/article/3678172/how-synthetic-media-will-transform-business-forever.html | access-date = 9 November 2022 }} increasing art-making accessibility, and artistic output per effort or expenses or time—e.g., via generating drafts, draft-definitions, and image components (inpainting). Generated images are sometimes used as sketches,{{cite news | last1 = Roose | first1 = Kevin | title = A.I.-Generated Art Is Already Transforming Creative Work | url = https://www.nytimes.com/2022/10/21/technology/ai-generated-art-jobs-dall-e-2.html | access-date = 16 November 2022 | work = The New York Times | date = 21 October 2022 }} low-cost experiments,{{cite news | last1 = Leswing | first1 = Kif | title = Why Silicon Valley is so excited about awkward drawings done by artificial intelligence | url = https://www.cnbc.com/2022/10/08/generative-ai-silicon-valleys-next-trillion-dollar-companies.html | access-date = 16 November 2022 | work = CNBC | language = en }} inspiration, or illustrations of proof-of-concept-stage ideas. Additional functionalities or improvements may also relate to post-generation manual editing (i.e., polishing), such as subsequent tweaking with an image editor.

{{see also|Prompt engineering#Text-to-image}}

File:Fooocus 2.5.5 screenshot showing the prompt section.webp]]

Prompts for some text-to-image models can also include images and keywords and configurable parameters, such as artistic style, which is often used via keyphrases like "in the style of [name of an artist]" in the prompt{{cite news | last1 = Robertson | first1 = Adi | date = 15 November 2022 | title = How DeviantArt is navigating the AI art minefield | work = The Verge | url = https://www.theverge.com/2022/11/15/23449036/deviantart-ai-art-dreamup-training-data-controversy | access-date = 16 November 2022 }} /or selection of a broad aesthetic/art style.{{cite news | last1 = Proulx | first1 = Natalie | date = September 2022 | title = Are A.I.-Generated Pictures Art? | work = The New York Times | url = https://www.nytimes.com/2022/09/16/learning/are-ai-generated-pictures-art.html | access-date = 16 November 2022 }} There are platforms for sharing, trading, searching, forking/refining, or collaborating on prompts for generating specific imagery from image generators.{{cite news | last1 = Vincent | first1 = James | date = 15 September 2022 | title = Anyone can use this AI art generator — that's the risk | work = The Verge | url = https://www.theverge.com/2022/9/15/23340673/ai-image-generation-stable-diffusion-explained-ethics-copyright-data | access-date = 9 November 2022 }}{{cite news | last1 = Davenport | first1 = Corbin | title = This AI Art Gallery Is Even Better Than Using a Generator | url = https://www.howtogeek.com/831697/this-ai-art-gallery-is-even-better-than-using-a-generator/ | access-date = 9 November 2022 | work = How-To Geek }}{{cite news | last1 = Robertson | first1 = Adi | title = Professional AI whisperers have launched a marketplace for DALL-E prompts | url = https://www.theverge.com/2022/9/2/23326868/dalle-midjourney-ai-promptbase-prompt-market-sales-artist-interview | access-date = 9 November 2022 | work = The Verge | date = 2 September 2022 }}{{cite news | title = Text-zu-Bild-Revolution: Stable Diffusion ermöglicht KI-Bildgenerieren für alle | url = https://www.heise.de/news/Text-zu-Bild-Revolution-Stable-Diffusion-ermoeglicht-KI-Bildgenerieren-fuer-alle-7244307.html | access-date = 9 November 2022 | work = heise online | language = de }} Prompts are often shared along with images on image-sharing websites such as Reddit and AI art-dedicated websites. A prompt is not the complete input needed for the generation of an image; additional inputs that determine the generated image include the output resolution, random seed, and random sampling parameters.{{Cite web | url = https://cdn.openart.ai/assets/Stable%20Diffusion%20Prompt%20Book%20From%20OpenArt%2011-13.pdf | title = Stable Diffusion Prompt Book | author = Mohamad Diab, Julian Herrera, Musical Sleep, Bob Chernow, Coco Mao | date = 2022-10-28 | access-date = 2023-08-07 }}

Synthetic media, which includes AI art, was described in 2022 as a major technology-driven trend that will affect business in the coming years. Harvard Kennedy School researchers voiced concerns about synthetic media serving as a vector for political misinformation soon after studying the proliferation of AI art on the X platform.{{Cite journal |last=Corsi |first=Giulio |last2=Marino |first2=Bill |last3=Wong |first3=Willow |date=2024-06-03 |title=The spread of synthetic media on X |url=https://misinforeview.hks.harvard.edu/article/the-spread-of-synthetic-media-on-x/ |journal=Harvard Kennedy School Misinformation Review |language=en-US |doi=10.37016/mr-2020-140|doi-access=free }} Synthography is a proposed term for the practice of generating images that are similar to photographs using AI.{{cite web | url = https://scholar.google.com/citations?view_op=view_citation&hl=en&user=cjLjVk8AAAAJ&citation_for_view=cjLjVk8AAAAJ:hC7cP41nSMkC | title = Synthography–An Invitation to Reconsider the Rapidly Changing Toolkit of Digital Image Creation as a New Genre Beyond Photography | last = Reinhuber | first = Elke | date = 2 December 2021 | publisher = Google Scholar | access-date = 20 December 2022 }}

{{Further|Algorithmic bias}}

A major concern raised about AI-generated images and art is sampling bias within model training data leading towards discriminatory output from AI art models. In 2023, University of Washington researchers found evidence of racial bias within the Stable Diffusion model, with images of a "person" corresponding most frequently with images of males from Europe or North America.{{Cite news |last=Milne |first=Stefan |date=November 29, 2023 |title=AI image generator Stable Diffusion perpetuates racial and gendered stereotypes, study finds |work=UW News |url=https://www.washington.edu/news/2023/11/29/ai-image-generator-stable-diffusion-perpetuates-racial-and-gendered-stereotypes-bias/}}

Looking more into the sampling bias found within AI training data, in 2017, researchers at Princeton University used AI software to link over 2 million words, finding that European names were viewed as more "pleasant" than African-Americans names, and that the words "woman" and "girl" were more likely to be associated with the arts instead of science and math, "which were most likely connected to males."{{Cite web |date=18 April 2017 |first=Adam |last=Hadhazy |title=Biased bots: Artificial-intelligence systems echo human prejudices |url=https://www.princeton.edu/news/2017/04/18/biased-bots-artificial-intelligence-systems-echo-human-prejudices |access-date=2024-11-13 |website=Office of Engineering Communications - Princeton University |language=en }} Generative AI models typically work based on user-entered word-based prompts, especially in the case of diffusion models, and this word-related bias may lead to biased results.

Along with this, generative AI can perpetuate harmful stereotypes regarding women. For example, Lensa, an AI app that trended on TikTok in 2023, was known to lighten black skin, make users thinner, and generate hypersexualized images of women.Fox, V. (March 11, 2023). AI Art & the Ethical Concerns of Artists. Beautiful Bizarre Magazine. Retrieved September 24, 2024, from https://beautifulbizarre.net/2023/03/11/ai-art-ethical-concerns-of-artists/ Melissa Heikkilä, a senior reporter at MIT Technology Review, shared the findings of an experiment using Lensa, noting that the generated avatars did not resemble her and often depicted her in a hypersexualized manner.{{Cite web |last=Heikkilä |first=Melissa |title=The viral AI avatar app Lensa undressed me—without my consent |url=https://www.technologyreview.com/2022/12/12/1064751/the-viral-ai-avatar-app-lensa-undressed-me-without-my-consent/ |access-date=2024-11-26 |website=MIT Technology Review |language=en}} Experts suggest that such outcomes can result from biases in the datasets used to train AI models, which can sometimes contain imbalanced representations, including hypersexual or nude imagery.{{Cite web |last=Lamensch |first=Marie |title=Generative AI Tools Are Perpetuating Harmful Gender Stereotypes |url=https://www.cigionline.org/articles/generative-ai-tools-are-perpetuating-harmful-gender-stereotypes/ |access-date=2024-11-26 |website=Centre for International Governance Innovation |language=en}}{{Cite book |last1=Birhane |first1=Abeba |last2=Prabhu |first2=Vinay Uday |chapter=Large image datasets: A pyrrhic win for computer vision? |date=1 July 2020 |title=2021 IEEE Winter Conference on Applications of Computer Vision (WACV) |pages= 1536–1546|doi=10.1109/WACV48630.2021.00158|arxiv=2006.16923 |isbn=978-1-6654-0477-8 |s2cid=220265500 }}

In 2024, Google's chatbot Gemini's AI image generator was criticized for perceived racial bias, with claims that Gemini deliberately underrepresented white people in its results.{{Cite web |last=Robertson |first=Adi |date=2024-02-21 |title=Google apologizes for "missing the mark" after Gemini generated racially diverse Nazis |url=https://www.theverge.com/2024/2/21/24079371/google-ai-gemini-generative-inaccurate-historical |url-status=live |archive-url=https://web.archive.org/web/20240421192422/https://www.theverge.com/2024/2/21/24079371/google-ai-gemini-generative-inaccurate-historical |archive-date=21 April 2024 |access-date=2024-04-20 |website=The Verge |language=en }} Users reported that it generated images of white historical figures like the Founding Fathers, Nazi soldiers, and Vikings as other races, and that it refused to process prompts such as "happy white people" and "ideal nuclear family".{{Cite web |last=Crimmins |first=Tricia |date=2024-02-21 |title=Why Google's new AI Gemini accused of refusing to acknowledge the existence of white people |url=https://www.dailydot.com/debug/google-ai-gemini-white-people/ |url-status=live |archive-url=https://web.archive.org/web/20240508175819/https://www.dailydot.com/debug/google-ai-gemini-white-people/ |archive-date=8 May 2024 |access-date=2024-05-08 |website=The Daily Dot |language=en-US }} Google later apologized for "missing the mark" and took Gemini's image generator offline for updates.{{Cite web |last=Raghavan |first=Prabhakar |date=2024-02-23 |title=Gemini image generation got it wrong. We'll do better. |url=https://blog.google/products/gemini/gemini-image-generation-issue/ |url-status=live |archive-url=https://web.archive.org/web/20240421033917/https://blog.google/products/gemini/gemini-image-generation-issue/ |archive-date=21 April 2024 |access-date=2024-04-20 |website=Google |language=en-us }} This prompted discussions about the ethical implications{{Cite web |date=2024-04-02 |title=Unmasking Racism in AI: From Gemini's Overcorrection to AAVE Bias and Ethical Considerations {{!}} Race & Social Justice Review |url=https://race-and-social-justice-review.law.miami.edu/unmasking-racism-in-ai-from-geminis-overcorrection-to-aave-bias-and-ethical-considerations/ |access-date=2024-10-26 |language=en-US}} of representing historical figures through a contemporary lens, leading critics to argue that these outputs could mislead audiences regarding actual historical contexts.{{Cite web |title=Rendering misrepresentation: Diversity failures in AI image generation |url=https://www.brookings.edu/articles/rendering-misrepresentation-diversity-failures-in-ai-image-generation/ |access-date=2024-10-26 |website=Brookings |language=en-US}} In addition to the well-documented representational issues such as racial and gender bias, some scholars have also pointed out deeper conceptual assumptions that shape how we perceive AI-generated art. For instance, framing AI strictly as a passive tool overlooks how cultural and technological factors influence its outputs. Others suggest viewing AI as part of a collaborative creative process, where both human and machine contribute to the artistic result.{{Cite journal |last=Tao |first=Feng |date=2022-03-04 |title=A New Harmonisation of Art and Technology: Philosophic Interpretations of Artificial Intelligence Art |url=https://www.tandfonline.com/doi/full/10.1080/02560046.2022.2112725 |journal=Critical Arts |language=en |volume=36 |issue=1-2 |pages=110–125 |doi=10.1080/02560046.2022.2112725 |issn=0256-0046}}

{{Further|Artificial intelligence and copyright}}

Legal scholars, artists, and media corporations have considered the legal and ethical implications of artificial intelligence art since the 20th century. Some artists use AI art to critique and explore the ethics of using gathered data to produce new artwork.{{Cite journal | last1 = Stark | first1 = Luke | last2 = Crawford | first2 = Kate | date = 2019-09-07 | title = The Work of Art in the Age of Artificial Intelligence: What Artists Can Teach Us About the Ethics of Data Practice | url = https://ojs.library.queensu.ca/index.php/surveillance-and-society/article/view/10821 | journal = Surveillance & Society | language = en | volume = 17 | issue = 3/4 | pages = 442–455 | doi = 10.24908/ss.v17i3/4.10821 | s2cid = 214218440 | issn = 1477-7487 | doi-access= free }}

In 1985, intellectual property law professor Pamela Samuelson argued that US copyright should allocate algorithmically generated artworks to the user of the computer program.{{Cite journal | last = Pamela | first = Samuelson | title = Allocating Ownership Rights in Computer-Generated Works | url = https://lawcat.berkeley.edu/record/1112407?ln=en | volume = 47 | journal = U. Pittsburgh L. Rev. | page = 1185 | year = 1985 }} A 2019 Florida Law Review article presented three perspectives on the issue. In the first, artificial intelligence itself would become the copyright owner; to do this, Section 101 of the US Copyright Act would need to be amended to define "author" as a computer. In the second, following Samuelson's argument, the user, programmer, or artificial intelligence company would be the copyright owner. This would be an expansion of the "work for hire" doctrine, under which ownership of a copyright is transferred to the "employer." In the third situation, copyright assignments would never take place, and such works would be in the public domain, as copyright assignments require an act of authorship.{{Cite journal | last = Victor | first = Palace | title = What if Artificial Intelligence Wrote This? Artificial Intelligence and Copyright Law | url = https://scholarship.law.ufl.edu/cgi/viewcontent.cgi?article=1439&context=flr | journal = Fla. L. Rev. | volume = 71 | issue = 1 | pages = 231–241 | date = January 2019 }}

In 2022, coinciding with the rising availability of consumer-grade AI image generation services, popular discussion renewed over the legality and ethics of AI-generated art. A particular topic is the inclusion of copyrighted artwork and images in AI training datasets, with artists objecting to commercial AI products using their works without consent, credit, or financial compensation.{{Cite magazine | last = Chayka | first = Kyle | date = 2023-02-10 | title = Is A.I. Art Stealing from Artists? | url = https://www.newyorker.com/culture/infinite-scroll/is-ai-art-stealing-from-artists | magazine = The New Yorker | language = en-US | issn = 0028-792X | access-date = 2023-09-06 }} In September 2022, Reema Selhi, of the Design and Artists Copyright Society, stated that "there are no safeguards for artists to be able to identify works in databases that are being used and opt out."{{cite web | author-last = Vallance | author-first = Chris | date = 13 September 2022 | title = "Art is dead Dude" - the rise of the AI artists stirs debate | url = https://www.bbc.com/news/technology-62788725 | access-date = 2 October 2022 | work = BBC News }} Some have claimed that images generated with these models can bear resemblance to extant artwork, sometimes including the remains of the original artist's signature.{{cite web | url = https://kotaku.com/ai-art-dall-e-midjourney-stable-diffusion-copyright-1849388060 | title = AI Creating 'Art' Is An Ethical And Copyright Nightmare | author-last = Plunkett | author-first = Luke | website = Kotaku | date = 25 August 2022 | access-date = 21 December 2022 }} In December 2022, users of the portfolio platform ArtStation staged an online protest against non-consensual use of their artwork within datasets; this resulted in opt-out services, such as "Have I Been Trained?" increasing in profile, as well as some online art platforms promising to offer their own opt-out options.{{cite web | url = https://arstechnica.com/information-technology/2022/12/artstation-artists-stage-mass-protest-against-ai-generated-artwork/ | title = Artists stage mass protest against AI-generated artwork on ArtStation | author-last = Edwards | author-first = Benj | website = Ars Technica | date = 15 December 2022 | access-date = 21 December 2022 }} According to the US Copyright Office, artificial intelligence programs are unable to hold copyright,{{Cite web | url = https://www.smithsonianmag.com/smart-news/us-copyright-office-rules-ai-art-cant-be-copyrighted-180979808/ | title = U.S. Copyright Office Rules A.I. Art Can't Be Copyrighted | first1 = Smithsonian | last1 = Magazine | first2 = Jane | last2 = Recker | website = Smithsonian Magazine }}{{Cite web | url = https://www.engadget.com/us-copyright-office-art-ai-creativity-machine-190722809.html | title = You can't copyright AI-created art, according to US officials | website = Engadget | date = 13 December 2022 }}{{Cite web | url = https://www.copyright.gov/rulings-filings/review-board/docs/a-recent-entrance-to-paradise.pdf | title = Re: Second Request for Reconsideration for Refusal to Register A Recent Entrance to Paradise }} a decision upheld at the Federal District level as of August 2023 followed the reasoning from the monkey selfie copyright dispute.{{cite web | url = https://www.hollywoodreporter.com/business/business-news/ai-works-not-copyrightable-studios-1235570316/ | title = AI-Created Art Isn't Copyrightable, Judge Says in Ruling That Could Give Hollywood Studios Pause | first = Winston | last = Cho | date = August 18, 2023 | access-date = August 19, 2023 | work = Hollywood Reporter }}

OpenAI, the developer of DALL-E, has its own policy on who owns generated art. They assign the right and title of a generated image to the creator, meaning the user who inputted the prompt owns the image generated, along with the right to sell, reprint, and merchandise it.Can I sell images I create with DALL·E? (n.d.). OpenAI Help Center. Retrieved November 11, 2024, from https://help.openai.com/en/articles/6425277-can-i-sell-images-i-create-with-dall-e

In January 2023, three artists—Sarah Andersen, Kelly McKernan, and Karla Ortiz—filed a copyright infringement lawsuit against Stability AI, Midjourney, and DeviantArt, claiming that it is legally required to obtain the consent of artists before training neural nets on their work and that these companies infringed on the rights of millions of artists by doing so on five billion images scraped from the web.[https://www.theverge.com/2023/1/16/23557098/generative-ai-art-copyright-legal-lawsuit-stable-diffusion-midjourney-deviantart James Vincent "AI art tools Stable Diffusion and Midjourney targeted with copyright lawsuit" The Verge, 16 January 2023.] In July 2023, U.S. District Judge William Orrick was inclined to dismiss most of the lawsuits filed by Andersen, McKernan, and Ortiz, but allowed them to file a new complaint.{{Cite news | last = Brittain | first = Blake | date = 2023-07-19 | title = US judge finds flaws in artists' lawsuit against AI companies | language = en | work = Reuters | url = https://www.reuters.com/legal/litigation/us-judge-finds-flaws-artists-lawsuit-against-ai-companies-2023-07-19/ | access-date = 2023-08-06 }} Also in 2023, Stability AI was sued by Getty Images for using its images in the training data.{{Cite web | last = Korn | first = Jennifer | date = 2023-01-17 | title = Getty Images suing the makers of popular AI art tool for allegedly stealing photos | url = https://www.cnn.com/2023/01/17/tech/getty-images-stability-ai-lawsuit/index.html | access-date = 2023-01-22 | website = CNN | language = en }} A tool built by Simon Willison allowed people to search 0.5% of the training data for Stable Diffusion V1.1, i.e., 12 million of the 2.3 billion instances from LAION 2B. Artist Karen Hallion discovered that her copyrighted images were used as training data without their consent.{{Cite book | last1 = Jiang | first1 = Harry H. | last2 = Brown | first2 = Lauren | last3 = Cheng | first3 = Jessica | last4 = Khan | first4 = Mehtab | last5 = Gupta | first5 = Abhishek | last6 = Workman | first6 = Deja | last7 = Hanna | first7 = Alex | last8 = Flowers | first8 = Johnathan | last9 = Gebru | first9 = Timnit | title = Proceedings of the 2023 AAAI/ACM Conference on AI, Ethics, and Society | chapter = AI Art and its Impact on Artists | date = 2023-08-08 | chapter-url = https://dl.acm.org/doi/10.1145/3600211.3604681 | language = en | publisher = ACM | pages = 363–374 | doi = 10.1145/3600211.3604681 | isbn = 979-8-4007-0231-0 | s2cid = 261279983 }}

In March 2024, Tennessee enacted the ELVIS Act, which prohibits the use of AI to mimic a musician's voice without permission.{{Cite web |last=Rosman |first=Rebecca |date=March 22, 2024 |title=Tennessee becomes the first state to protect musicians and other artists against AI |website=NPR |url=https://www.npr.org/2024/03/22/1240114159/tennessee-protect-musicians-artists-ai}} A month later in that year, Adam Schiff introduced the Generative AI Copyright Disclosure Act which, if passed, would require that AI companies to submit copyrighted works in their datasets to the Register of Copyrights before releasing new generative AI systems.{{Cite web |last=Robins-Early |first=Nick |date=April 9, 2024 |title=New bill would force AI companies to reveal use of copyrighted art {{!}} Artificial intelligence (AI) {{!}} The Guardian |url=https://www.theguardian.com/technology/2024/apr/09/artificial-intelligence-bill-copyright-art |access-date=2024-04-13 |website=amp.theguardian.com}} In November 2024, a group of artists and activists shared early access to OpenAI’s unreleased video generation model, Sora, via Huggingface. The action, accompanied by a statement, criticized the exploitative use of artists’ work by major corporations.'{{Cite journal |last=Elwes |first=Jake |last2=CROSSLUCID |last3=Vettier |first3=Aurèce |last4=Rauh |first4=Maribeth |date=2024-11-26 |title=Art in the Cage of Digital Reproduction |url=https://artinthecageofdigitalreproduction.org/ |journal=Art in the Cage of Digital Reproduction |language=en |publisher=Art in the Cage Collective}}{{Cite news |last=Murgia |first=Madhumita |last2=Criddle |first2=Cristina |date=2024-11-26 |title=OpenAI’s text-to-video AI tool Sora leaked in protest by artists |url=https://www.ft.com/content/5281eff4-711b-49ac-8227-634dbeed757b |access-date=2025-02-07 |work=Financial Times}}{{Cite web |last=Spangler |first=Todd |date=2024-11-27 |title=OpenAI Shuts Down Sora Access After Artists Released Video-Generation Tool in Protest: ‘We Are Not Your PR Puppets’ |url=https://variety.com/2024/digital/news/openai-shuts-down-sora-artists-protest-leak-1236224878/ |access-date=2025-02-07 |website=Variety |language=en-US}}

As with other types of photo manipulation since the early 19th century, some people in the early 21st century have been concerned that AI could be used to create content that is misleading and can be made to damage a person's reputation, such as deepfakes.{{cite news | last1 = Wiggers | first1 = Kyle | date = 24 August 2022 | title = Deepfakes: Uncensored AI art model prompts ethics questions | work = TechCrunch | url = https://techcrunch.com/2022/08/24/deepfakes-for-all-uncensored-ai-art-model-prompts-ethics-questions/ | access-date = 15 September 2022 }} Artist Sarah Andersen, who previously had her art copied and edited to depict Neo-Nazi beliefs, stated that the spread of hate speech online can be worsened by the use of image generators. Some also generate images or videos for the purpose of catfishing.

AI systems have the ability to create deepfake content, which is often viewed as harmful and offensive. The creation of deepfakes poses a risk to individuals who have not consented to it. This mainly refers to deepfake pornography which is used as revenge porn, where sexually explicit material is disseminated to humiliate or harm another person. AI-generated child pornography has been deemed a potential danger to society due to its unlawful nature.{{cite web |last=Beahm |first=Anna |date=12 February 2024 |title=What you need to know about the ongoing fight to prevent AI-generated child porn |url=https://www.reckon.news/news/2024/02/what-you-need-to-know-about-the-ongoing-fight-to-prevent-ai-generated-child-porn.html |url-status=live |archive-url=https://web.archive.org/web/20240307181355/https://www.reckon.news/news/2024/02/what-you-need-to-know-about-the-ongoing-fight-to-prevent-ai-generated-child-porn.html#:~:text=Most%20AI%20CSAM%20found%20is,still%20harm%20children%2C%20IWF%20said |archive-date=7 March 2024 |website=Reckon News |access-date=7 March 2024 }}

File:EldagsenElectrician.jpg|Pseudomnesia: The Electrician won Boris Eldagsen one of the categories in the Sony World Photography Awards competition.

File:Pope Francis in puffy winter jacket.jpg|A 2023 AI-generated image of Pope Francis wearing a puffy winter jacket fooled some viewers into believing it was an actual photograph. It went viral on social media platforms.

File:Trump’s arrest (2).jpg|Journalist Eliot Higgins' Midjourney-generated image depicts former President Donald Trump getting arrested. The image was posted on Twitter and went viral.{{Cite web |last=Higgins |first=Eliot |author-link=Eliot Higgins |date=March 21, 2023 |title=Making pictures of Trump getting arrested while waiting for Trump's arrest. |url=https://twitter.com/EliotHiggins/status/1637927681734987777 |url-status=dead |archive-url=https://web.archive.org/web/20230420191543/https://twitter.com/EliotHiggins/status/1637927681734987777 |archive-date=April 20, 2023 |via=Twitter }}

File:AI generated figure published in a Frontiers journal.png|One of the seven AI-generated images that were used for figures in the now-retracted paper Cellular functions of spermatogonial stem cells in relation to JAK/STAT signaling pathway. Figure 1, "Spermatogonial stem cells, isolated, purified and cultured from rat testes".

After winning the 2023 "Creative" "Open competition" Sony World Photography Awards, Boris Eldagsen stated that his entry was actually created with artificial intelligence. Photographer Feroz Khan commented to the BBC that Eldagsen had "clearly shown that even experienced photographers and art experts can be fooled".{{cite news | title = Sony World Photography Award 2023: Winner refuses award after revealing AI creation | url = https://www.bbc.com/news/entertainment-arts-65296763 | access-date = 16 June 2023 | work = BBC News | date = 17 April 2023 }} Smaller contests have been affected as well; in 2023, a contest run by author Mark Lawrence as Self-Published Fantasy Blog-Off was cancelled after the winning entry was allegedly exposed to be a collage of images generated with Midjourney.{{cite news | last1 = Sato | first1 = Mia | title = How AI art killed an indie book cover contest | url = https://www.theverge.com/2023/6/9/23752354/ai-spfbo-cover-art-contest-midjourney-clarkesworld | access-date = 19 June 2023 | work = The Verge | date = 9 June 2023 }}

In May 2023, on social media sites such as Reddit and Twitter, attention was given to a Midjourney-generated image of Pope Francis wearing a white puffer coat.{{Cite web | last = Novak | first = Matt | title = That Viral Image Of Pope Francis Wearing A White Puffer Coat Is Totally Fake | url = https://www.forbes.com/sites/mattnovak/2023/03/26/that-viral-image-of-pope-francis-wearing-a-white-puffer-coat-is-totally-fake/ | access-date = 2023-06-16 | website = Forbes | language = en }}{{Cite web | last = Stokel-Walker | first = Chris | date = 2023-03-27 | title = We Spoke To The Guy Who Created The Viral AI Image Of The Pope That Fooled The World | url = https://www.buzzfeednews.com/article/chrisstokelwalker/pope-puffy-jacket-ai-midjourney-image-creator-interview | access-date = 2023-06-16 | website = BuzzFeed News | language = en }} Additionally, an AI-generated image of an attack on the Pentagon went viral as part of a hoax news story on Twitter.{{cite web | last=Edwards | first=Benj | title=Fake Pentagon "explosion" photo sows confusion on Twitter | website=Ars Technica | date=2023-05-23 | url=https://arstechnica.com/information-technology/2023/05/ai-generated-image-of-explosion-near-pentagon-goes-viral-sparks-brief-panic/ | access-date=2024-07-02}}{{cite news | last1=Oremus | first1=Will | last2=Harwell | first2=Drew | last3=Armus | first3=Teo | title=A tweet about a Pentagon explosion was fake. It still went viral. | newspaper=Washington Post | date=2023-05-22 | url=https://www.washingtonpost.com/technology/2023/05/22/pentagon-explosion-ai-image-hoax/ | access-date=2024-07-02}}

In the days before March 2023 indictment of Donald Trump as part of the Stormy Daniels–Donald Trump scandal, several AI-generated images allegedly depicting Trump's arrest went viral online.{{Cite news |last1=Devlin |first1=Kayleen |last2=Cheetham |first2=Joshua |date=2023-03-25 |title=Fake Trump arrest photos: How to spot an AI-generated image |url=https://www.bbc.com/news/world-us-canada-65069316 |url-status=live |archive-url=https://web.archive.org/web/20240412054703/https://www.bbc.com/news/world-us-canada-65069316 |archive-date=12 April 2024 |access-date=2024-02-24 |language=en-GB}}{{Cite web | date = 2023-03-24 | title = Trump shares deepfake photo of himself praying as AI images of arrest spread online | url = https://www.independent.co.uk/news/world/americas/us-politics/donald-trump-ai-praying-photo-b2307178.html | access-date = 2023-06-16 | website = The Independent | language = en }} On March 20, British journalist Eliot Higgins generated various images of Donald Trump being arrested or imprisoned using Midjourney v5 and posted them on Twitter; two images of Trump struggling against arresting officers went viral under the mistaken impression that they were genuine, accruing more than 5 million views in three days.{{Cite web |last=Garber |first=Megan |date=2023-03-24 |title=The Trump AI Deepfakes Had an Unintended Side Effect |url=https://www.theatlantic.com/culture/archive/2023/03/fake-trump-arrest-images-ai-generated-deepfakes/673510/ |url-status=live |archive-url=https://web.archive.org/web/20240518213950/https://www.theatlantic.com/culture/archive/2023/03/fake-trump-arrest-images-ai-generated-deepfakes/673510/ |archive-date=18 May 2024 |access-date=2024-04-21 |website=The Atlantic |language=en }}{{Cite web |last=Lasarte |first=Diego |date=2023-03-23 |title=As fake photos of Trump's "arrest" went viral, Trump shared an AI-generated photo too |url=https://qz.com/trump-ai-photo-arrest-truthsocial-twitter-1850259197 |url-status=live |archive-url=https://web.archive.org/web/20240421133139/https://qz.com/trump-ai-photo-arrest-truthsocial-twitter-1850259197 |archive-date=21 April 2024 |access-date=2024-04-21 |website=Quartz (publication) |language=en }} According to Higgins, the images were not meant to mislead, but he was banned from using Midjourney services as a result. As of April 2024, the tweet had garnered more than 6.8 million views.

In February 2024, the paper Cellular functions of spermatogonial stem cells in relation to JAK/STAT signaling pathway was published using AI-generated images. It was later retracted from Frontiers in Cell and Developmental Biology because the paper "does not meet the standards".{{Cite journal |last1=Guo |first1=Xinyu |last2=Dong |first2=Liang |last3=Hao |first3=Dingjun |date=2024 |editor-last=Kumaresan |editor-first=Arumugam |title=Cellular functions of spermatogonial stem cells in relation to JAK/STAT signaling pathway |journal=Frontiers in Cell and Developmental Biology |volume=12 |doi=10.3389/fcell.2024.1386861 |issn=2296-634X |doi-access=free}}

To mitigate some deceptions, OpenAI developed a tool in 2024 to detect images that were generated by DALL-E 3.{{Cite web |last=Whitwam |first=Ryan |date=May 8, 2024 |title=New OpenAI Tool Can Detect Dall-E 3 AI Images With 98% Accuracy |url=https://www.extremetech.com/computing/new-openai-tool-can-detect-dall-e-3-ai-images-with-98-accuracy |url-status=live |archive-url=https://web.archive.org/web/20240526050013/https://www.extremetech.com/computing/new-openai-tool-can-detect-dall-e-3-ai-images-with-98-accuracy |archive-date=26 May 2024 |access-date=May 26, 2024 |website=ExtremeTech}} In testing, this tool accurately identified DALL-E 3-generated images approximately 98% of the time. The tool is also fairly capable of recognizing images that have been visually modified by users post-generation.{{Cite web |date=7 May 2024 |title=OpenAI's new tool can detect images created by DALL-E 3 |url=https://www.fastcompany.com/91120099/openai-new-tool-detect-images-created-by-dall-e-3}}

{{Further|Workplace impact of artificial intelligence|Technological unemployment}}

File:Théâtre D’opéra Spatial.png, an image generated with Midjourney which won a digital art competition in 2022]]

As generative AI image software such as Stable Diffusion and DALL-E continue to advance, the potential problems and concerns that these systems pose for creativity and artistry have risen. In 2022, artists working in various media raised concerns about the impact that generative artificial intelligence could have on their ability to earn money, particularly if AI-based images started replacing artists working in the illustration and design industries.{{cite web | author-last = King | author-first = Hope | date = 10 August 2022 | title = AI-generated digital art spurs debate about news illustrations | url = https://www.axios.com/2022/08/10/artificial-intelligence-digital-art-journalism | access-date = 2 October 2022 | work = Axios }}{{cite web | author-last = Salkowitz | author-first = Rob | date = 16 September 2022 | title = AI Is Coming For Commercial Art Jobs. Can It Be Stopped? | url = https://www.forbes.com/sites/robsalkowitz/2022/09/16/ai-is-coming-for-commercial-art-jobs-can-it-be-stopped/?sh=5e6784c654b0 | access-date = 2 October 2022 | work = Forbes }} In August 2022, digital artist R. J. Palmer stated that "I could easily envision a scenario where using AI, a single artist or art director could take the place of 5–10 entry level artists... I have seen a lot of self-published authors and such say how great it will be that they don’t have to hire an artist." Scholars Jiang et al. state that "Leaders of companies like Open AI and Stability AI have openly stated that they expect generative AI systems to replace creatives imminently." A 2022 case study found that AI-produced images created by technology like DALL-E caused some traditional artists to be concerned about losing work, while others use it to their advantage and view it as a tool.{{Cite news |last=Parra |first=Dex |date=February 24, 2023 |title=CASE STUDY: The Case of DALLE-2 |url=https://mediaengagement.org/research/the-ethics-of-ai-art/ |work=University of Texas at Austin, Center for Media Management}}

AI-based images have become more commonplace in art markets and search engines because AI-based text-to-image systems are trained from pre-existing artistic images, sometimes without the original artist's consent, allowing the software to mimic specific artists' styles.{{Cite book | last1 = Inie | first1 = Nanna | last2 = Falk | first2 = Jeanette | last3 = Tanimoto | first3 = Steve | title = Extended Abstracts of the 2023 CHI Conference on Human Factors in Computing Systems | chapter = Designing Participatory AI: Creative Professionals' Worries and Expectations about Generative AI | date = 2023-04-19 | chapter-url = https://dl.acm.org/doi/10.1145/3544549.3585657 | language = en | publisher = ACM | pages = 1–8 | doi = 10.1145/3544549.3585657 | arxiv = 2303.08931 | isbn = 978-1-4503-9422-2 | s2cid = 257557305 }} For example, Polish digital artist Greg Rutkowski has stated that it is more difficult to search for his work online because many of the images in the results are AI-generated specifically to mimic his style. Furthermore, some training databases on which AI systems are based are not accessible to the public.

The ability of AI-based art software to mimic or forge artistic style also raises concerns of malice or greed.{{cite web | last1 = Roose | first1 = Kevin | year = 2022 | title = An A.I.-Generated Picture Won an Art Prize. Artists Aren't Happy. | url = https://www.nytimes.com/2022/09/02/technology/ai-artificial-intelligence-artists.html | work = The New York Times }}{{cite news | title = An AI-Generated Artwork Won First Place at a State Fair Fine Arts Competition, and Artists Are Pissed | language = en | work = Vice | url = https://www.vice.com/en/article/bvmvqm/an-ai-generated-artwork-won-first-place-at-a-state-fair-fine-arts-competition-and-artists-are-pissed | access-date = 15 September 2022 }} Works of AI-generated art, such as Théâtre D'opéra Spatial, a text-to-image AI illustration that won the grand prize in the August 2022 digital art competition at the Colorado State Fair, have begun to overwhelm art contests and other submission forums meant for small artists. The Netflix short film The Dog & the Boy, released in January 2023, received backlash online for its use of artificial intelligence art to create the film's background artwork.{{cite web |last1=Chen |first1=Min |date=7 February 2023 |title=Netflix Japan Is Drawing Ire for Using A.I. to Generate the Background Art of Its New Anime Short |url=https://news.artnet.com/news/netflix-japan-ai-anime-dog-and-boy-2251247 |url-status=live |archive-url=https://web.archive.org/web/20231202020450/https://news.artnet.com/news/netflix-japan-ai-anime-dog-and-boy-2251247 |archive-date=2 December 2023 |access-date=2 December 2023 |website=Artnet }} Within the same vein, Disney released Secret Invasion, a Marvel TV show with an AI-generated intro, on Disney+ in 2023, causing concern and backlash regarding the idea that artists could be made obsolete by machine-learning tools.Pulliam, C. (2023, June 27). Marvel’s Secret Invasion AI credits should shock no one. The Verge. Retrieved August 26, 2024, from https://www.theverge.com/2023/6/27/23770133/secret-invasion-ai-credits-marvel

AI art has sometimes been deemed to be able to replace traditional stock images.{{Cite web |last=Tolliver-Walker |first=Heidi |date=October 11, 2023 |title=Can AI-Generated Images Replace Stock? |url=https://whattheythink.com/articles/116873-can-ai-generated-images-replace-stock/ |access-date=2024-05-26 |website=WhatTheyThink |language=en}} In 2023, Shutterstock announced a beta test of an AI tool that can regenerate partial content of other Shutterstock's images. Getty Images and Nvidia have partnered with the launch of Generative AI by iStock, a model trained on Getty's library and iStock's photo library using Nvidia's Picasso model.{{Cite web |last=David |first=Emilia |date=2024-01-08 |title=Getty and Nvidia bring generative AI to stock photos |url=https://www.theverge.com/2024/1/8/24027259/getty-images-nvidia-generative-ai-stock-photos |url-status=live |archive-url=https://web.archive.org/web/20240526043817/https://www.theverge.com/2024/1/8/24027259/getty-images-nvidia-generative-ai-stock-photos |archive-date=26 May 2024 |access-date=2024-05-26 |website=The Verge |language=en }}

File:Cartoon dynamo.jpg humorously imagines the life of a cartoonist in 2023, when machines powered by electricity can produce and execute ideas for cartoons.]]

Researchers from Hugging Face and Carnegie Mellon University reported in a 2023 paper that generating one thousand 1024×1024 images using Stable Diffusion's XL 1.0 base model requires 11.49 kWh of energy and generates {{Convert|1,594|g}} of carbon dioxide, which is roughly equivalent to driving an average gas-powered car a distance of {{convert|4.1|miles}}. Comparing 88 different models, the paper concluded that image-generation models used on average around 2.9{{Nbsp}}kWh of energy per 1,000 inferences.{{cite book |last1=Luccioni |first1=Alexandra Sasha |author-link1=Sasha Luccioni|title=The 2024 ACM Conference on Fairness, Accountability, and Transparency |last2=Jernite |first2=Yacine |last3=Strubell |first3=Emma |date=2024 |isbn=979-8-4007-0450-5 |pages=85–99 |chapter=Power Hungry Processing: Watts Driving the Cost of AI Deployment? |doi=10.1145/3630106.3658542 |arxiv=2311.16863}}

In addition to the creation of original art, research methods that use AI have been generated to quantitatively analyze digital art collections. This has been made possible due to the large-scale digitization of artwork in the past few decades. According to CETINIC and SHE (2022), using artificial intelligence to analyze already-existing art collections can provide new perspectives on the development of artistic styles and the identification of artistic influences.{{Cite journal | last1 = Cetinic | first1 = Eva | last2 = She | first2 = James | date = 2022-05-31 | title = Understanding and Creating Art with AI: Review and Outlook | url = https://dl.acm.org/doi/10.1145/3475799 | journal = ACM Transactions on Multimedia Computing, Communications, and Applications | language = en | volume = 18 | issue = 2 | pages = 1–22 | arxiv = 2102.09109 | doi = 10.1145/3475799 | issn = 1551-6857 | s2cid = 231951381 }}{{Cite journal | last1 = Cetinic | first1 = Eva | last2 = She | first2 = James | date = 2022-02-16 | title = Understanding and Creating Art with AI: Review and Outlook | journal = ACM Transactions on Multimedia Computing, Communications, and Applications | volume = 18 | issue = 2 | pages = 66:1–66Kate Vass2 | arxiv = 2102.09109 | doi = 10.1145/3475799 | issn = 1551-6857 | s2cid = 231951381 }}

Two computational methods, close reading and distant viewing, are the typical approaches used to analyze digitized art.{{Cite conference | last1 = Lang | first1 = Sabine | last2 = Ommer | first2 = Bjorn | date = 2018 | title = Reflecting on How Artworks Are Processed and Analyzed by Computer Vision: Supplementary Material | url = https://openaccess.thecvf.com/content_eccv_2018_workshops/w13/html/Lang_Reflecting_on_How_Artworks_Are_Processed_and_Analyzed_by_Computer_ECCVW_2018_paper.html | via = Computer Vision Foundation | book-title = Proceedings of the European Conference on Computer Vision (ECCV) Workshops }} Close reading focuses on specific visual aspects of one piece. Some tasks performed by machines in close reading methods include computational artist authentication and analysis of brushstrokes or texture properties. In contrast, through distant viewing methods, the similarity across an entire collection for a specific feature can be statistically visualized. Common tasks relating to this method include automatic classification, object detection, multimodal tasks, knowledge discovery in art history, and computational aesthetics. Synthetic images can also be used to train AI algorithms for art authentication and to detect forgeries.{{Cite journal |last1=Ostmeyer |first1=Johann |last2=Schaerf |first2=Ludovica |last3=Buividovich |first3=Pavel |last4=Charles |first4=Tessa |last5=Postma |first5=Eric |last6=Popovici |first6=Carina |date=2024-02-14 |title=Synthetic images aid the recognition of human-made art forgeries |journal=PLOS ONE |publication-place=United States |volume=19 |issue=2 |pages=e0295967 |arxiv=2312.14998 |doi=10.1371/journal.pone.0295967 |issn=1932-6203 |pmc=10866502 |pmid=38354162 |doi-access=free|bibcode=2024PLoSO..1995967O }}

Researchers have also introduced models that predict emotional responses to art. One such model is ArtEmis, a large-scale dataset paired with machine learning models. ArtEmis includes emotional annotations from over 6,500 participants along with textual explanations. By analyzing both visual inputs and the accompanying text descriptions from this dataset, ArtEmis enables the generation of nuanced emotional predictions.{{Cite arXiv |eprint=2101.07396 |class=cs.CV |first1=Panos |last1=Achlioptas |first2=Maks |last2=Ovsjanikov |title=ArtEmis: Affective Language for Visual Art |date=2021-01-18 |last3=Haydarov |first3=Kilichbek |last4=Elhoseiny |first4=Mohamed |last5=Guibas |first5=Leonidas}}{{Cite web |last=Myers |first=Andrew |date=2021-03-22 |title=Artist's Intent: AI Recognizes Emotions in Visual Art |url=https://hai.stanford.edu/news/artists-intent-ai-recognizes-emotions-visual-art |access-date=2024-11-24 |website=hai.stanford.edu |language=en}}

AI has also been used in arts outside of visual arts. Generative AI has been used in video game production beyond imagery, especially for level design (e.g., for custom maps) and creating new content (e.g., quests or dialogue) or interactive stories in video games.{{cite book | last1 = Yannakakis | first1 = Geogios N. | title = Proceedings of the 9th conference on Computing Frontiers | date = 15 May 2012 | isbn = 9781450312158 | pages = 285–292 | chapter = Game AI revisited | doi = 10.1145/2212908.2212954 | s2cid = 4335529 }}{{cite news | date = 8 May 2018 | title = AI creates new levels for Doom and Super Mario games | work = BBC News | url = https://www.bbc.com/news/technology-44040007 | access-date = 9 November 2022 }} AI has also been used in the literary arts,{{cite journal | last1 = Katsnelson | first1 = Alla | date = 29 August 2022 | title = Poor English skills? New AIs help researchers to write better | journal = Nature | language = en | volume = 609 | issue = 7925 | pages = 208–209 | bibcode = 2022Natur.609..208K | doi = 10.1038/d41586-022-02767-9 | pmid = 36038730 | s2cid = 251931306 | doi-access = free }} such as helping with writer's block, inspiration, or rewriting segments.{{cite web | last1 = | first1 = | date = 9 November 2022 | title = KoboldAI/KoboldAI-Client | url = https://github.com/KoboldAI/KoboldAI-Client | access-date = 9 November 2022 | website = GitHub }}{{cite news | last1 = Dzieza | first1 = Josh | date = 20 July 2022 | title = Can AI write good novels? | work = The Verge | url = https://www.theverge.com/c/23194235/ai-fiction-writing-amazon-kindle-sudowrite-jasper | access-date = 16 November 2022 }}{{cite news | title = AI Writing Assistants: A Cure for Writer's Block or Modern-Day Clippy? | language = en | work = PCMAG | url = https://www.pcmag.com/how-to/ai-writing-assistants-a-cure-for-writers-block-or-modern-day-clippy | access-date = 16 November 2022 }}{{cite news | last1 = Song | first1 = Victoria | date = 2 November 2022 | title = Google's new prototype AI tool does the writing for you | work = The Verge | url = https://www.theverge.com/2022/11/2/23435258/google-ai-writing-wordcraft-lamda | access-date = 16 November 2022 }} In the culinary arts, some prototype cooking robots can dynamically taste, which can assist chefs in analyzing the content and flavor of dishes during the cooking process.{{cite journal | last1 = Sochacki | first1 = Grzegorz | last2 = Abdulali | first2 = Arsen | last3 = Iida | first3 = Fumiya | date = 2022 | title = Mastication-Enhanced Taste-Based Classification of Multi-Ingredient Dishes for Robotic Cooking | journal = Frontiers in Robotics and AI | volume = 9 | page = 886074 | doi = 10.3389/frobt.2022.886074 | issn = 2296-9144 | pmc = 9114309 | pmid = 35603082 | doi-access = free }}

{{Columns-list|* Algorithmic art

• AI slop
• {{section link|Applications of artificial intelligence|Art}}
• {{annotated link|Artificial intelligence and elections}}
• Artificial intelligence in architecture
• Computational creativity
• Cybernetic art
• Generative art
• List of artificial intelligence artists
• Music and artificial intelligence
• Neural style transfer
• Synthetic media}}

{{reflist}}

{{Digital art}}

{{Western art movements}}

Category:20th-century introductions

Category:20th-century art movements

Category:Generative artificial intelligence

Category:Visual arts

Category:Digital art

Category:Computer art

Category:Art controversies

Category:Works involved in plagiarism controversies

Category:Articles containing video clips