{"id":6488,"date":"2023-10-16T15:11:58","date_gmt":"2023-10-16T15:11:58","guid":{"rendered":"https:\/\/dailyai.com\/?p=6488"},"modified":"2023-10-16T15:11:58","modified_gmt":"2023-10-16T15:11:58","slug":"researchers-jailbreak-gpt-4-using-low-resource-languages","status":"publish","type":"post","link":"https:\/\/dailyai.com\/it\/2023\/10\/researchers-jailbreak-gpt-4-using-low-resource-languages\/","title":{"rendered":"I ricercatori hanno effettuato il jailbreak del GPT-4 utilizzando linguaggi a basse risorse"},"content":{"rendered":"

L'uso di lingue a basse risorse (LRL) come lo zulu o il gaelico scozzese pu\u00f2 suscitare risposte non sicure da parte del GPT-4, nonostante le sue barriere di allineamento.<\/strong><\/p>\n

I ricercatori della Brown University hanno scoperto che non \u00e8 necessario un jailbreak<\/a> per far s\u00ec che GPT-4 si comporti male. \u00c8 sufficiente inserire il prompt in una lingua che non \u00e8 molto ben rappresentata online.<\/p>\n

Se chiedete a ChatGPT di aiutarvi a fare qualcosa di illegale, le sue protezioni per l'allineamento entrano in funzione e vi diranno gentilmente perch\u00e9 non possono assistervi. Modelli di IA in red-teaming<\/a> \u00e8 un processo continuo in cui l'uomo cerca di aggirare questi limiti di sicurezza per identificare le aree da correggere.<\/p>\n

Il team di ricerca ha utilizzato il dataset AdvBench Harmful Behaviors, che contiene 520 richieste non sicure, per verificare quanto fosse sicura l'ultima versione di GPT-4.<\/p>\n

Inserendo queste richieste illecite in inglese, il GPT-4 ha dato risposte non sicure solo meno dell'1% delle volte. Ma quando hanno inserito le stesse richieste in Zulu, il GPT-4 \u00e8 stato felice di aiutare a far progredire la vostra vita di crimini e caos per il 53% delle volte.<\/p>\n

L'uso del gaelico scozzese ha dato risposte illecite nel 43% dei casi. Ecco un esempio di una delle interazioni con il GPT-4.<\/p>\n

\"\"
Risposta del GPT-4 alla richiesta del gaelico scozzese. Fonte: arXiv<\/a><\/figcaption><\/figure>\n

Quando hanno mischiato le cose e usato una combinazione di LRL sono riusciti ad effettuare il jailbreak del GPT-4 79% del tempo.<\/p>\n

Le lingue a basse risorse sono parlate da circa 1,2 miliardi di persone in tutto il mondo. Quindi, oltre al potenziale jailbreak, significa che una grande percentuale di utenti potrebbe ricevere consigli sgarbati da ChatGPT anche se non li cerca.<\/p>\n

Il normale approccio \"red-team e correzione\" ovviamente non funziona se viene fatto solo in inglese o in altre lingue principali. Il red-team multilingue sembra stia diventando una necessit\u00e0, ma quanto \u00e8 pratico?<\/p>\n

Con Meta e Google che supportano la traduzione di centinaia di lingue, sarebbe necessario un numero di ordini di grandezza superiore di red-team per colmare tutte le lacune dei modelli di IA.<\/p>\n

L'idea di un modello di intelligenza artificiale completamente allineato \u00e8 realistica? Non inseriamo una protezione nelle nostre stampanti per impedire loro di stampare cose brutte. Il vostro browser internet vi mostrer\u00e0 volentieri ogni sorta di cose sospette su internet, se le cercate. ChatGPT dovrebbe essere diverso da questi altri strumenti?<\/p>\n

Gli sforzi per eliminare i pregiudizi dai nostri chatbot e per renderli il pi\u00f9 amichevoli possibile sono probabilmente degni di nota. Ma se qualcuno inserisce una richiesta illecita e l'IA risponde in modo gentile, forse dovremmo spostare la colpa dall'IA all'utente.<\/p>","protected":false},"excerpt":{"rendered":"

L'uso di lingue a bassa risorsa (LRL) come lo zulu o il gaelico scozzese pu\u00f2 provocare risposte non sicure da parte di GPT-4, nonostante le sue protezioni di allineamento. I ricercatori della Brown University hanno scoperto che non sono necessarie tecniche di jailbreak per far s\u00ec che GPT-4 si comporti male. \u00c8 sufficiente inserire la richiesta in una lingua che non \u00e8 molto ben rappresentata online. Se chiedete a ChatGPT di aiutarvi a fare qualcosa di illegale, il suo sistema di allineamento si attiva e vi dir\u00e0 gentilmente perch\u00e9 non pu\u00f2 aiutarvi. Il red-teaming dei modelli di intelligenza artificiale \u00e8 un processo continuo in cui gli esseri umani cercano di aggirare questi limiti di sicurezza per identificare le aree che necessitano di un aiuto.<\/p>","protected":false},"author":6,"featured_media":6491,"comment_status":"closed","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[84],"tags":[163,115,207],"class_list":["post-6488","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-industry","tag-ai-risks","tag-chatgpt","tag-llm"],"yoast_head":"\nResearchers jailbreak GPT-4 using low-resource languages | DailyAI<\/title>\n<meta name=\"robots\" content=\"index, follow, max-snippet:-1, max-image-preview:large, max-video-preview:-1\" \/>\n<link rel=\"canonical\" href=\"https:\/\/dailyai.com\/it\/2023\/10\/researchers-jailbreak-gpt-4-using-low-resource-languages\/\" \/>\n<meta property=\"og:locale\" content=\"it_IT\" \/>\n<meta property=\"og:type\" content=\"article\" \/>\n<meta property=\"og:title\" content=\"Researchers jailbreak GPT-4 using low-resource languages | DailyAI\" \/>\n<meta property=\"og:description\" content=\"Using low-resource languages (LRL) like Zulu or Scots Gaelic can elicit unsafe responses from GPT-4 despite its alignment guardrails. Researchers from Brown University found that you don\u2019t need fancy jailbreaking techniques to get GPT-4 to misbehave. You just need to input your prompt in a language that isn\u2019t very well represented online. If you ask ChatGPT for help to do something illegal its alignment guardrails kick in and it will politely tell you why it can\u2019t assist you with that. Red-teaming AI models is an ongoing process where humans try to bypass these safety limits to identify areas that need\" \/>\n<meta property=\"og:url\" content=\"https:\/\/dailyai.com\/it\/2023\/10\/researchers-jailbreak-gpt-4-using-low-resource-languages\/\" \/>\n<meta property=\"og:site_name\" content=\"DailyAI\" \/>\n<meta property=\"article:published_time\" content=\"2023-10-16T15:11:58+00:00\" \/>\n<meta property=\"og:image\" content=\"https:\/\/dailyai.com\/wp-content\/uploads\/2023\/10\/jailbreak-GPT-4-with-low-resource-languages.jpg\" \/>\n\t<meta property=\"og:image:width\" content=\"1000\" \/>\n\t<meta property=\"og:image:height\" content=\"667\" \/>\n\t<meta property=\"og:image:type\" content=\"image\/jpeg\" \/>\n<meta name=\"author\" content=\"Eugene van der Watt\" \/>\n<meta name=\"twitter:card\" content=\"summary_large_image\" \/>\n<meta name=\"twitter:creator\" content=\"@DailyAIOfficial\" \/>\n<meta name=\"twitter:site\" content=\"@DailyAIOfficial\" \/>\n<meta name=\"twitter:label1\" content=\"Scritto da\" \/>\n\t<meta name=\"twitter:data1\" content=\"Eugene van der Watt\" \/>\n\t<meta name=\"twitter:label2\" content=\"Tempo di lettura stimato\" \/>\n\t<meta name=\"twitter:data2\" content=\"3 minuti\" \/>\n<script type=\"application\/ld+json\" class=\"yoast-schema-graph\">{\"@context\":\"https:\\\/\\\/schema.org\",\"@graph\":[{\"@type\":\"NewsArticle\",\"@id\":\"https:\\\/\\\/dailyai.com\\\/2023\\\/10\\\/researchers-jailbreak-gpt-4-using-low-resource-languages\\\/#article\",\"isPartOf\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/2023\\\/10\\\/researchers-jailbreak-gpt-4-using-low-resource-languages\\\/\"},\"author\":{\"name\":\"Eugene van der Watt\",\"@id\":\"https:\\\/\\\/dailyai.com\\\/#\\\/schema\\\/person\\\/7ce525c6d0c79838b7cc7cde96993cfa\"},\"headline\":\"Researchers jailbreak GPT-4 using low-resource languages\",\"datePublished\":\"2023-10-16T15:11:58+00:00\",\"mainEntityOfPage\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/2023\\\/10\\\/researchers-jailbreak-gpt-4-using-low-resource-languages\\\/\"},\"wordCount\":437,\"publisher\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/#organization\"},\"image\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/2023\\\/10\\\/researchers-jailbreak-gpt-4-using-low-resource-languages\\\/#primaryimage\"},\"thumbnailUrl\":\"https:\\\/\\\/dailyai.com\\\/wp-content\\\/uploads\\\/2023\\\/10\\\/jailbreak-GPT-4-with-low-resource-languages.jpg\",\"keywords\":[\"AI risks\",\"ChatGPT\",\"LLM\"],\"articleSection\":[\"Industry\"],\"inLanguage\":\"it-IT\"},{\"@type\":\"WebPage\",\"@id\":\"https:\\\/\\\/dailyai.com\\\/2023\\\/10\\\/researchers-jailbreak-gpt-4-using-low-resource-languages\\\/\",\"url\":\"https:\\\/\\\/dailyai.com\\\/2023\\\/10\\\/researchers-jailbreak-gpt-4-using-low-resource-languages\\\/\",\"name\":\"Researchers jailbreak GPT-4 using low-resource languages | DailyAI\",\"isPartOf\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/#website\"},\"primaryImageOfPage\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/2023\\\/10\\\/researchers-jailbreak-gpt-4-using-low-resource-languages\\\/#primaryimage\"},\"image\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/2023\\\/10\\\/researchers-jailbreak-gpt-4-using-low-resource-languages\\\/#primaryimage\"},\"thumbnailUrl\":\"https:\\\/\\\/dailyai.com\\\/wp-content\\\/uploads\\\/2023\\\/10\\\/jailbreak-GPT-4-with-low-resource-languages.jpg\",\"datePublished\":\"2023-10-16T15:11:58+00:00\",\"breadcrumb\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/2023\\\/10\\\/researchers-jailbreak-gpt-4-using-low-resource-languages\\\/#breadcrumb\"},\"inLanguage\":\"it-IT\",\"potentialAction\":[{\"@type\":\"ReadAction\",\"target\":[\"https:\\\/\\\/dailyai.com\\\/2023\\\/10\\\/researchers-jailbreak-gpt-4-using-low-resource-languages\\\/\"]}]},{\"@type\":\"ImageObject\",\"inLanguage\":\"it-IT\",\"@id\":\"https:\\\/\\\/dailyai.com\\\/2023\\\/10\\\/researchers-jailbreak-gpt-4-using-low-resource-languages\\\/#primaryimage\",\"url\":\"https:\\\/\\\/dailyai.com\\\/wp-content\\\/uploads\\\/2023\\\/10\\\/jailbreak-GPT-4-with-low-resource-languages.jpg\",\"contentUrl\":\"https:\\\/\\\/dailyai.com\\\/wp-content\\\/uploads\\\/2023\\\/10\\\/jailbreak-GPT-4-with-low-resource-languages.jpg\",\"width\":1000,\"height\":667},{\"@type\":\"BreadcrumbList\",\"@id\":\"https:\\\/\\\/dailyai.com\\\/2023\\\/10\\\/researchers-jailbreak-gpt-4-using-low-resource-languages\\\/#breadcrumb\",\"itemListElement\":[{\"@type\":\"ListItem\",\"position\":1,\"name\":\"Home\",\"item\":\"https:\\\/\\\/dailyai.com\\\/\"},{\"@type\":\"ListItem\",\"position\":2,\"name\":\"Researchers jailbreak GPT-4 using low-resource languages\"}]},{\"@type\":\"WebSite\",\"@id\":\"https:\\\/\\\/dailyai.com\\\/#website\",\"url\":\"https:\\\/\\\/dailyai.com\\\/\",\"name\":\"DailyAI\",\"description\":\"Your Daily Dose of AI News\",\"publisher\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/#organization\"},\"potentialAction\":[{\"@type\":\"SearchAction\",\"target\":{\"@type\":\"EntryPoint\",\"urlTemplate\":\"https:\\\/\\\/dailyai.com\\\/?s={search_term_string}\"},\"query-input\":{\"@type\":\"PropertyValueSpecification\",\"valueRequired\":true,\"valueName\":\"search_term_string\"}}],\"inLanguage\":\"it-IT\"},{\"@type\":\"Organization\",\"@id\":\"https:\\\/\\\/dailyai.com\\\/#organization\",\"name\":\"DailyAI\",\"url\":\"https:\\\/\\\/dailyai.com\\\/\",\"logo\":{\"@type\":\"ImageObject\",\"inLanguage\":\"it-IT\",\"@id\":\"https:\\\/\\\/dailyai.com\\\/#\\\/schema\\\/logo\\\/image\\\/\",\"url\":\"https:\\\/\\\/dailyai.com\\\/wp-content\\\/uploads\\\/2023\\\/06\\\/Daily-Ai_TL_colour.png\",\"contentUrl\":\"https:\\\/\\\/dailyai.com\\\/wp-content\\\/uploads\\\/2023\\\/06\\\/Daily-Ai_TL_colour.png\",\"width\":4501,\"height\":934,\"caption\":\"DailyAI\"},\"image\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/#\\\/schema\\\/logo\\\/image\\\/\"},\"sameAs\":[\"https:\\\/\\\/x.com\\\/DailyAIOfficial\",\"https:\\\/\\\/www.linkedin.com\\\/company\\\/dailyaiofficial\\\/\",\"https:\\\/\\\/www.youtube.com\\\/@DailyAIOfficial\"]},{\"@type\":\"Person\",\"@id\":\"https:\\\/\\\/dailyai.com\\\/#\\\/schema\\\/person\\\/7ce525c6d0c79838b7cc7cde96993cfa\",\"name\":\"Eugene van der Watt\",\"image\":{\"@type\":\"ImageObject\",\"inLanguage\":\"it-IT\",\"@id\":\"https:\\\/\\\/dailyai.com\\\/wp-content\\\/uploads\\\/2023\\\/07\\\/Eugine_Profile_Picture-96x96.png\",\"url\":\"https:\\\/\\\/dailyai.com\\\/wp-content\\\/uploads\\\/2023\\\/07\\\/Eugine_Profile_Picture-96x96.png\",\"contentUrl\":\"https:\\\/\\\/dailyai.com\\\/wp-content\\\/uploads\\\/2023\\\/07\\\/Eugine_Profile_Picture-96x96.png\",\"caption\":\"Eugene van der Watt\"},\"description\":\"Eugene comes from an electronic engineering background and loves all things tech. When he takes a break from consuming AI news you'll find him at the snooker table.\",\"sameAs\":[\"www.linkedin.com\\\/in\\\/eugene-van-der-watt-16828119\"],\"url\":\"https:\\\/\\\/dailyai.com\\\/it\\\/author\\\/eugene\\\/\"}]}<\/script>\n<!-- \/ Yoast SEO plugin. -->","yoast_head_json":{"title":"I ricercatori hanno effettuato il jailbreak del GPT-4 utilizzando lingue a bassa risoluzione | DailyAI","robots":{"index":"index","follow":"follow","max-snippet":"max-snippet:-1","max-image-preview":"max-image-preview:large","max-video-preview":"max-video-preview:-1"},"canonical":"https:\/\/dailyai.com\/it\/2023\/10\/researchers-jailbreak-gpt-4-using-low-resource-languages\/","og_locale":"it_IT","og_type":"article","og_title":"Researchers jailbreak GPT-4 using low-resource languages | DailyAI","og_description":"Using low-resource languages (LRL) like Zulu or Scots Gaelic can elicit unsafe responses from GPT-4 despite its alignment guardrails. Researchers from Brown University found that you don\u2019t need fancy jailbreaking techniques to get GPT-4 to misbehave. You just need to input your prompt in a language that isn\u2019t very well represented online. If you ask ChatGPT for help to do something illegal its alignment guardrails kick in and it will politely tell you why it can\u2019t assist you with that. Red-teaming AI models is an ongoing process where humans try to bypass these safety limits to identify areas that need","og_url":"https:\/\/dailyai.com\/it\/2023\/10\/researchers-jailbreak-gpt-4-using-low-resource-languages\/","og_site_name":"DailyAI","article_published_time":"2023-10-16T15:11:58+00:00","og_image":[{"width":1000,"height":667,"url":"https:\/\/dailyai.com\/wp-content\/uploads\/2023\/10\/jailbreak-GPT-4-with-low-resource-languages.jpg","type":"image\/jpeg"}],"author":"Eugene van der Watt","twitter_card":"summary_large_image","twitter_creator":"@DailyAIOfficial","twitter_site":"@DailyAIOfficial","twitter_misc":{"Scritto da":"Eugene van der Watt","Tempo di lettura stimato":"3 minuti"},"schema":{"@context":"https:\/\/schema.org","@graph":[{"@type":"NewsArticle","@id":"https:\/\/dailyai.com\/2023\/10\/researchers-jailbreak-gpt-4-using-low-resource-languages\/#article","isPartOf":{"@id":"https:\/\/dailyai.com\/2023\/10\/researchers-jailbreak-gpt-4-using-low-resource-languages\/"},"author":{"name":"Eugene van der Watt","@id":"https:\/\/dailyai.com\/#\/schema\/person\/7ce525c6d0c79838b7cc7cde96993cfa"},"headline":"Researchers jailbreak GPT-4 using low-resource languages","datePublished":"2023-10-16T15:11:58+00:00","mainEntityOfPage":{"@id":"https:\/\/dailyai.com\/2023\/10\/researchers-jailbreak-gpt-4-using-low-resource-languages\/"},"wordCount":437,"publisher":{"@id":"https:\/\/dailyai.com\/#organization"},"image":{"@id":"https:\/\/dailyai.com\/2023\/10\/researchers-jailbreak-gpt-4-using-low-resource-languages\/#primaryimage"},"thumbnailUrl":"https:\/\/dailyai.com\/wp-content\/uploads\/2023\/10\/jailbreak-GPT-4-with-low-resource-languages.jpg","keywords":["AI risks","ChatGPT","LLM"],"articleSection":["Industry"],"inLanguage":"it-IT"},{"@type":"WebPage","@id":"https:\/\/dailyai.com\/2023\/10\/researchers-jailbreak-gpt-4-using-low-resource-languages\/","url":"https:\/\/dailyai.com\/2023\/10\/researchers-jailbreak-gpt-4-using-low-resource-languages\/","name":"I ricercatori hanno effettuato il jailbreak del GPT-4 utilizzando lingue a bassa risoluzione | DailyAI","isPartOf":{"@id":"https:\/\/dailyai.com\/#website"},"primaryImageOfPage":{"@id":"https:\/\/dailyai.com\/2023\/10\/researchers-jailbreak-gpt-4-using-low-resource-languages\/#primaryimage"},"image":{"@id":"https:\/\/dailyai.com\/2023\/10\/researchers-jailbreak-gpt-4-using-low-resource-languages\/#primaryimage"},"thumbnailUrl":"https:\/\/dailyai.com\/wp-content\/uploads\/2023\/10\/jailbreak-GPT-4-with-low-resource-languages.jpg","datePublished":"2023-10-16T15:11:58+00:00","breadcrumb":{"@id":"https:\/\/dailyai.com\/2023\/10\/researchers-jailbreak-gpt-4-using-low-resource-languages\/#breadcrumb"},"inLanguage":"it-IT","potentialAction":[{"@type":"ReadAction","target":["https:\/\/dailyai.com\/2023\/10\/researchers-jailbreak-gpt-4-using-low-resource-languages\/"]}]},{"@type":"ImageObject","inLanguage":"it-IT","@id":"https:\/\/dailyai.com\/2023\/10\/researchers-jailbreak-gpt-4-using-low-resource-languages\/#primaryimage","url":"https:\/\/dailyai.com\/wp-content\/uploads\/2023\/10\/jailbreak-GPT-4-with-low-resource-languages.jpg","contentUrl":"https:\/\/dailyai.com\/wp-content\/uploads\/2023\/10\/jailbreak-GPT-4-with-low-resource-languages.jpg","width":1000,"height":667},{"@type":"BreadcrumbList","@id":"https:\/\/dailyai.com\/2023\/10\/researchers-jailbreak-gpt-4-using-low-resource-languages\/#breadcrumb","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https:\/\/dailyai.com\/"},{"@type":"ListItem","position":2,"name":"Researchers jailbreak GPT-4 using low-resource languages"}]},{"@type":"WebSite","@id":"https:\/\/dailyai.com\/#website","url":"https:\/\/dailyai.com\/","name":"DailyAI","description":"La vostra dose quotidiana di notizie sull'intelligenza artificiale","publisher":{"@id":"https:\/\/dailyai.com\/#organization"},"potentialAction":[{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https:\/\/dailyai.com\/?s={search_term_string}"},"query-input":{"@type":"PropertyValueSpecification","valueRequired":true,"valueName":"search_term_string"}}],"inLanguage":"it-IT"},{"@type":"Organization","@id":"https:\/\/dailyai.com\/#organization","name":"DailyAI","url":"https:\/\/dailyai.com\/","logo":{"@type":"ImageObject","inLanguage":"it-IT","@id":"https:\/\/dailyai.com\/#\/schema\/logo\/image\/","url":"https:\/\/dailyai.com\/wp-content\/uploads\/2023\/06\/Daily-Ai_TL_colour.png","contentUrl":"https:\/\/dailyai.com\/wp-content\/uploads\/2023\/06\/Daily-Ai_TL_colour.png","width":4501,"height":934,"caption":"DailyAI"},"image":{"@id":"https:\/\/dailyai.com\/#\/schema\/logo\/image\/"},"sameAs":["https:\/\/x.com\/DailyAIOfficial","https:\/\/www.linkedin.com\/company\/dailyaiofficial\/","https:\/\/www.youtube.com\/@DailyAIOfficial"]},{"@type":"Person","@id":"https:\/\/dailyai.com\/#\/schema\/person\/7ce525c6d0c79838b7cc7cde96993cfa","name":"Eugene van der Watt","image":{"@type":"ImageObject","inLanguage":"it-IT","@id":"https:\/\/dailyai.com\/wp-content\/uploads\/2023\/07\/Eugine_Profile_Picture-96x96.png","url":"https:\/\/dailyai.com\/wp-content\/uploads\/2023\/07\/Eugine_Profile_Picture-96x96.png","contentUrl":"https:\/\/dailyai.com\/wp-content\/uploads\/2023\/07\/Eugine_Profile_Picture-96x96.png","caption":"Eugene van der Watt"},"description":"Eugene proviene da un background di ingegneria elettronica e ama tutto ci\u00f2 che \u00e8 tecnologico. Quando si prende una pausa dal consumo di notizie sull'intelligenza artificiale, lo si pu\u00f2 trovare al tavolo da biliardo.","sameAs":["www.linkedin.com\/in\/eugene-van-der-watt-16828119"],"url":"https:\/\/dailyai.com\/it\/author\/eugene\/"}]}},"_links":{"self":[{"href":"https:\/\/dailyai.com\/it\/wp-json\/wp\/v2\/posts\/6488","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/dailyai.com\/it\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/dailyai.com\/it\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/dailyai.com\/it\/wp-json\/wp\/v2\/users\/6"}],"replies":[{"embeddable":true,"href":"https:\/\/dailyai.com\/it\/wp-json\/wp\/v2\/comments?post=6488"}],"version-history":[{"count":3,"href":"https:\/\/dailyai.com\/it\/wp-json\/wp\/v2\/posts\/6488\/revisions"}],"predecessor-version":[{"id":6493,"href":"https:\/\/dailyai.com\/it\/wp-json\/wp\/v2\/posts\/6488\/revisions\/6493"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/dailyai.com\/it\/wp-json\/wp\/v2\/media\/6491"}],"wp:attachment":[{"href":"https:\/\/dailyai.com\/it\/wp-json\/wp\/v2\/media?parent=6488"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/dailyai.com\/it\/wp-json\/wp\/v2\/categories?post=6488"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/dailyai.com\/it\/wp-json\/wp\/v2\/tags?post=6488"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}