{"id":4584,"date":"2024-02-28T09:36:59","date_gmt":"2024-02-28T01:36:59","guid":{"rendered":"https:\/\/www.1ai.net\/?p=4584"},"modified":"2024-02-28T09:36:59","modified_gmt":"2024-02-28T01:36:59","slug":"%e6%8a%a5%e5%91%8a%ef%bc%9a60%e7%9a%84gpt-3-5%e8%be%93%e5%87%ba%e5%ad%98%e5%9c%a8%e6%8a%84%e8%a2%ad%e9%97%ae%e9%a2%98","status":"publish","type":"post","link":"https:\/\/www.1ai.net\/en\/4584.html","title":{"rendered":"Report: 60%&#039;s GPT-3.5 output has plagiarism issues"},"content":{"rendered":"<p>According to a report from Copyleaks, OpenAI\u2019s<a href=\"https:\/\/www.1ai.net\/en\/tag\/gpt\" title=\"_OTHER ORGANISER\" target=\"_blank\" >GPT<\/a>-3.5 Model output, 60% exists<a href=\"https:\/\/www.1ai.net\/en\/tag\/%e6%8a%84%e8%a2%ad\" title=\"[Sees articles containing [copy] labels]\" target=\"_blank\" >Plagiarism<\/a>Copyleaks uses a proprietary scoring method that takes into account factors such as identical text, minor edits and paraphrases, assigning each output a \u201csimilarity score\u201d.<\/p>\n<p class=\"article-content__img\"><img loading=\"lazy\" decoding=\"async\" class=\"alignnone size-full wp-image-4585\" title=\"202005261144375677_13\" src=\"https:\/\/www.1ai.net\/wp-content\/uploads\/2024\/02\/202005261144375677_13.jpg\" alt=\"202005261144375677_13\" width=\"600\" height=\"371\" \/><\/p>\n<p>GPT-3.5 is an advanced natural language processing model launched by OpenAI, but the originality of its output has been questioned.<span class=\"spamTxt\">up to date<\/span>The results of the study showed that among the outputs of GPT-3.5, 45.71 TP3T of the text was the same, 27.41 TP3T was slightly modified, and 46.51 TP3T was rewritten. A similarity score of 0.1 TP3T indicates completely original, while<span class=\"spamTxt\">100%<\/span>This means there is no original content.<\/p>\n<p>Copyleaks ran a variety of tests on GPT-3.5, generating about a thousand outputs in 26 disciplines, each with about 400 words. The results showed that the similarity score for computer science<span class=\"spamTxt\">Highest<\/span>\uff08<span class=\"spamTxt\">100%<\/span>), followed by physics (92%) and psychology (88%). In contrast, similarity scores for drama (0.9%), humanities (2.8%) and English language (5.4%) were<span class=\"spamTxt\">lowest<\/span>.<\/p>\n<p>\u201cOur models are designed and trained to learn concepts that help them solve new problems,\u201d said Lindsey Held, a spokesperson for OpenAI. \u201cWe have taken steps to limit incidental memorization, and our terms of use prohibit intentional use of our models to regurgitate content.\u201d<\/p>\n<p>The issue of plagiarism is not just about copying and pasting entire sentences or paragraphs. The New York Times once filed a lawsuit against OpenAI, claiming that the &quot;mass copying&quot; of OpenAI&#039;s AI system constituted copyright infringement. OpenAI responded by saying that &quot;occasional memory&quot; was a &quot;rare error&quot; and accused the New York Times of &quot;manipulating cues.&quot;<\/p>\n<p>While content creators, from authors to visual artists, have been arguing in court that the underlying technology that generates AI is trained on their copyrighted works, the law currently favors companies over plaintiffs. The New York Times case may offer a glimmer of hope, but progress is still pending.<\/p>","protected":false},"excerpt":{"rendered":"<p>According to a report by Copyleaks, 60% of OpenAI's GPT-3.5 model outputs were plagiarized.Copyleaks used a proprietary scoring methodology that takes into account identical text, minor modifications, and rewrites, and assigns a \"similarity score\" to each output. \". GPT-3.5 is an advanced natural language processing model from OpenAI, but the originality of its output has been questioned. According to the latest findings, 45.71 TP3T of GPT-3.5's output has the same text, 27.41 TP3T has been slightly modified, and 46.51 TP3T is rewritten text. A similarity score of 0% indicates complete originality, while 100% indicates no original content. Copyleaks' analysis of the GPT-3.5<\/p>","protected":false},"author":1,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[146],"tags":[271,1400],"collection":[],"class_list":["post-4584","post","type-post","status-publish","format-standard","hentry","category-news","tag-gpt","tag-1400"],"acf":[],"_links":{"self":[{"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/posts\/4584","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/comments?post=4584"}],"version-history":[{"count":0,"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/posts\/4584\/revisions"}],"wp:attachment":[{"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/media?parent=4584"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/categories?post=4584"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/tags?post=4584"},{"taxonomy":"collection","embeddable":true,"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/collection?post=4584"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}