{"id":359925,"date":"2025-12-20T12:35:08","date_gmt":"2025-12-20T12:35:08","guid":{"rendered":"https:\/\/www.newsbeep.com\/us\/359925\/"},"modified":"2025-12-20T12:35:08","modified_gmt":"2025-12-20T12:35:08","slug":"what-if-readers-like-a-i-generated-fiction","status":"publish","type":"post","link":"https:\/\/www.newsbeep.com\/us\/359925\/","title":{"rendered":"What if Readers Like A.I.-Generated Fiction?"},"content":{"rendered":"

After feeding Han\u2019s writing into GPT-4o, Chakrabarty fine-tuned fresh versions of the model on the work of twenty-nine other authors, including a close college friend of mine, Tony Tulathimutte. Jia Tolentino once praised<\/a> Tony\u2019s short stories, saying that his \u201cdeviant instincts crackle in nearly every line.\u201d I\u2019d been reading him since the early two-thousands\u2014and yet his A.I. clone could have easily fooled me. Here\u2019s a sample A.I.-generated line: \u201cHe finally counted 18 breaths, and, to delay longer, opened up a new doc and composed the marriage proposal he\u2019d send to the first man to make him cum without dildos or videos.\u201d<\/p>\n

Chakrabarty had started his project out of intellectual curiosity, but he was growing disturbed by its implications. Pangram, an A.I.-detection program, failed to flag almost all of the prose generated by his fine-tuned models. This suggested that anyone with some storytelling skills could feed a plot into a fine-tuned chatbot, slap their name on the resulting manuscript, and try to publish it. People often minimize A.I.-generated literature\u2014after all, we read books to access someone else\u2019s consciousness. But what if we can\u2019t tell the difference? When Chakrabarty returned from Japan, he invited Jane Ginsburg, a Columbia professor who specializes in copyright law, to join him and Dhillon as a co-author of a paper about the research. Ginsburg agreed. \u201cI don\u2019t know whether what I\u2019m scared about is the ability to produce this content,\u201d she told me, \u201cor the prospect that this content could be really commercially viable.\u201d<\/p>\n

Chakrabarty, now a computer-science professor at Stony Brook University, recently released<\/a> a preprint of the research, which has not yet been peer-reviewed. The paper notes that graduate students ultimately compared thirty A.I.-generated passages\u2014one imitating each author in the study\u2014with passages written by their colleagues. They weren\u2019t told what they were reading; they were simply asked which they liked best. They preferred the quality of the A.I. output in almost two-thirds of the cases.<\/p>\n

Reading the authors\u2019 original passages<\/a> alongside the A.I. imitations, I was startled to find that I liked some of the imitations just as much. The A.I. version of Han\u2019s scene, about the newborn\u2019s death, struck me as trite in places. But, to me, the line about the mother\u2019s chant was more surprising and exact than the original. I also spotted some good bits in an imitation of Junot D\u00edaz<\/a>. In \u201cThis Is How You Lose Her<\/a>,\u201d D\u00edaz writes, \u201cThe one thing she warned you about, that she swore she would never forgive, was cheating. I\u2019ll put a machete in you, she promised.\u201d To my ear, the A.I. rendition was more rhythmic and economical: \u201cShe told you from the beginning that if you ever cheated on her she would chop your little pito off.\u201d I\u2019d been studying Spanish for a couple of years, but I had to look up pito\u2014a word for \u201cwhistle\u201d that I hadn\u2019t heard before. Google\u2019s A.I. overview told me that, in some places, it was also slang for \u201cpenis.\u201d D\u00edazian enough, I figured.<\/p>\n

When I wrote to the authors whose work was used in the study, most declined to be interviewed or didn\u2019t respond. But a few e-mailed their thoughts. Lydia Davis wrote, \u201cI think the point is certainly made, that AI can create a decent paragraph that might deceive one into thinking it was written by a certain human being.\u201d Orhan Pamuk<\/a> said, \u201cI am sure soon there will be much more exact imitations.\u201d<\/p>\n

D\u00edaz and Sigrid Nunez<\/a> agreed to be interviewed. Over Zoom, I asked D\u00edaz about chopping someone\u2019s pito off. \u201cPito, of course, just means \u2018whistle,\u2019\u00a0\u201d he said, apparently perplexed. I told him that, according to the internet, it could also be a double-entendre. \u201cMy memory sucks, but, in all my years as a fucking Domincian in the diaspora, that is not a thing that I have ever heard,\u201d he told me. He thought that his doppelg\u00e4nger\u2019s vernacular was geographically and historically incoherent. \u201cI tend to write in a very specific time-stamped Jersey slang,\u201d he said. Plus, he added, the A.I.\u2019s rhythm and characterization were no good.<\/p>\n

Nunez described her A.I. copycat as \u201ccompletely banal.\u201d \u201cIt isn\u2019t my style, my story, my sensibility, my philosophy of life\u2014it\u2019s not me,\u201d she told me. \u201cIt\u2019s a machine that thinks that\u2019s what I\u2019m like.\u201d When I pointed out that skilled graduate students had found the passage well written, she questioned whether they had paid close enough attention, suggesting that they\u2019d made thoughtless judgments so that they could return to their own writing. (She didn\u2019t like their imitations, either.) \u201cIf I thought this reflected anything that actually had to do with my work, I\u2019d shoot myself,\u201d Nunez said.<\/p>\n","protected":false},"excerpt":{"rendered":"After feeding Han\u2019s writing into GPT-4o, Chakrabarty fine-tuned fresh versions of the model on the work of twenty-nine…\n","protected":false},"author":2,"featured_media":359926,"comment_status":"","ping_status":"","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[45],"tags":[3357,182,177780,181,507,62500,28946,74,177781,177779,777],"class_list":{"0":"post-359925","1":"post","2":"type-post","3":"status-publish","4":"format-standard","5":"has-post-thumbnail","7":"category-artificial-intelligence","8":"tag-_sensitivecontent","9":"tag-ai","10":"tag-annals-of-artificial-intelligence","11":"tag-artificial-intelligence","12":"tag-artificialintelligence","13":"tag-inverted","14":"tag-onecolumnnarrow","15":"tag-technology","16":"tag-textabovecenterfullbleednocontributor","17":"tag-the-weekend-essay","18":"tag-web"},"_links":{"self":[{"href":"https:\/\/www.newsbeep.com\/us\/wp-json\/wp\/v2\/posts\/359925","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.newsbeep.com\/us\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.newsbeep.com\/us\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.newsbeep.com\/us\/wp-json\/wp\/v2\/users\/2"}],"replies":[{"embeddable":true,"href":"https:\/\/www.newsbeep.com\/us\/wp-json\/wp\/v2\/comments?post=359925"}],"version-history":[{"count":0,"href":"https:\/\/www.newsbeep.com\/us\/wp-json\/wp\/v2\/posts\/359925\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.newsbeep.com\/us\/wp-json\/wp\/v2\/media\/359926"}],"wp:attachment":[{"href":"https:\/\/www.newsbeep.com\/us\/wp-json\/wp\/v2\/media?parent=359925"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.newsbeep.com\/us\/wp-json\/wp\/v2\/categories?post=359925"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.newsbeep.com\/us\/wp-json\/wp\/v2\/tags?post=359925"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}