At least 11,500 creative professionals \u2014 including Oscar-winning actor Julianne Moore, author James Patterson and Radiohead’s musician Thom Yorke \u2014 have signed an open letter calling for the prohibition of using human art to train artificial intelligence without permission.<\/p>\n\n\n\n
\u201cThe unlicensed use of creative works for training generative AI is a major, unjust threat to the livelihoods of the people behind those works, and must not be permitted,\u201d the one-line statement<\/a>, which isn’t addressed to anyone in particular, reads.<\/p>\n\n\n\n The lack of federal regulation around generative AI amid its rapid proliferation has contributed to a growing maelstrom of confusion and criticism around what datasets AI companies are using to train their models. Artists and companies in recent years have called out these tech giants and startups for allegedly scraping their work without their knowledge<\/a> or consent.<\/p>\n\n\n\n Ed Newton-Rex, a composer and former executive at Stability AI, said this movement for regulation comes at a critical time for creatives worldwide as legal battles and legislation over unlicensed AI training continue to unfold.<\/p>\n\n\n\n \u201cLots of generative AI companies train on creators\u2019 work without a license to do so,\u201d said Newton-Rex, who founded Fairly Trained, a nonprofit that certifies generative AI companies that use consensual training data. \u201cThis is a major issue for many of the artists, musicians, actors, authors and other creators whose work is being exploited by AI companies.\u201d<\/p>\n\n\n\n The letter has racked up hundreds more signatories in the hours since it went public on Tuesday.<\/p>\n\n\n\n Actor Fran Drescher, president of SAG-AFTRA, the labor union that represents thousands of Hollywood actors, was among the signatories. Dozens of organizations, including major publishing houses and music labels, signed the letter as well.<\/p>\n\n\n\n The potential for generative AI to replace human labor remains a pressing concern for actors<\/a> and video game performers<\/a>, as well as others in the industry.<\/p>\n\n\n\n Publishers like The New York Times<\/a> and The Wall Street Journal<\/a> have sued AI companies such as OpenAI and Perplexity AI for copyright infringement. Others, like Cond\u00e9 Nast<\/a> and Hearst<\/a>, have instead struck content deals with OpenAI \u2014 allowing its products, including ChatGPT and SearchGPT, to display content from the companies\u2019 media outlets.<\/p>\n\n\n\n The world\u2019s biggest record labels also teamed up in June<\/a> to take two prominent AI music-making companies to court, alleging that they trained their models on decades\u2019 worth of copyrighted sound recordings without consent.<\/p>\n\n\n\n The proliferation of accessible AI tools capable of generating realistic media \u2014 including full songs<\/a>, music videos<\/a> and advertisements<\/a> \u2014 has triggered a slew of legal and ethical questions for artists. Many artists have expressed concern<\/a> over how generative AI technologies could undermine human work and compensation, including by replicating celebrities\u2019 likenesses without their consent.<\/p>\n\n\n\n Earlier this year, Drake removed a diss track from the internet after Tupac Shakur\u2019s estate threatened to sue<\/a> him for allegedly using the late rapper\u2019s AI-generated voice in the song. AI-generated advertisements have also used the likenesses of public figures such as Taylor Swift<\/a>, MrBeast<\/a>, Scarlett Johansson<\/a> and Morgan Freeman<\/a> to promote various products.<\/p>\n\n\n\n For now, no comprehensive legislation exists in the United States to regulate the development of AI, but several states have begun introducing bills to propose specific regulations, particularly around deepfakes<\/a>. Last month, California passed into law<\/a> two bills aimed at protecting actors and other performers from unauthorized use of their digital likenesses.<\/p>\n\n\n\n