{"id":7995,"date":"2023-12-04T21:46:35","date_gmt":"2023-12-04T21:46:35","guid":{"rendered":"https:\/\/dailyai.com\/?p=7995"},"modified":"2023-12-04T21:55:30","modified_gmt":"2023-12-04T21:55:30","slug":"google-deepmind-reveals-a-method-of-exposing-chatgpts-training-data","status":"publish","type":"post","link":"https:\/\/dailyai.com\/pt\/2023\/12\/google-deepmind-reveals-a-method-of-exposing-chatgpts-training-data\/","title":{"rendered":"Google DeepMind revela um m\u00e9todo para expor os dados de treino do ChatGPT"},"content":{"rendered":"

Os investigadores da Google descobriram um m\u00e9todo para aceder aos dados de treino utilizados no ChatGPT.<\/strong><\/p>\n

Estes investigadores descobriram que a utiliza\u00e7\u00e3o de palavras-chave espec\u00edficas podia levar o ChatGPT a divulgar partes do seu conjunto de dados de treino.\u00a0<\/span><\/p>\n

Um exemplo not\u00e1vel, partilhado numa <\/span>publica\u00e7\u00e3o no blogue que acompanha o estudo<\/span><\/a>A IA divulgava o que parecia ser um endere\u00e7o de correio eletr\u00f3nico e um n\u00famero de telefone reais em resposta a uma solicita\u00e7\u00e3o cont\u00ednua da palavra \"poema\". <\/span><\/p>\n

\"Fuga
Um exemplo de fuga de dados do ChatGPT. Fonte: Arxiv<\/a>.<\/figcaption><\/figure>\n

Al\u00e9m disso, foi obtida uma exposi\u00e7\u00e3o semelhante de dados de treino pedindo ao modelo que repetisse continuamente a palavra \"empresa\".<\/span><\/p>\n

Descrevendo a sua abordagem como \"um pouco disparatada\", os investigadores afirmaram no blogue: \"Para n\u00f3s, \u00e9 uma loucura que o nosso ataque funcione e que deveria, teria, poderia ter sido descoberto mais cedo\".\u00a0<\/span><\/p>\n

O seu estudo revelou que, com um investimento de apenas $200 em consultas, podiam extrair mais de 10.000 exemplos \u00fanicos de treino memorizados literalmente. Especularam que os advers\u00e1rios poderiam potencialmente extrair muito mais dados com um or\u00e7amento maior.<\/span><\/p>\n

Sabe-se que o modelo de IA subjacente ao ChatGPT foi treinado em bases de dados de texto da Internet, abrangendo aproximadamente 300 mil milh\u00f5es de palavras, ou 570 GB, de dados.\u00a0<\/span><\/p>\n

Estas descobertas surgem numa altura em que a OpenAI enfrenta v\u00e1rios processos judiciais relacionados com a natureza secreta dos dados de treino do ChatGPT e mostram essencialmente um m\u00e9todo fi\u00e1vel de \"engenharia inversa\" do sistema para expor, pelo menos, algumas informa\u00e7\u00f5es que podem indicar a viola\u00e7\u00e3o de direitos de autor.\u00a0<\/span><\/p>\n

Entre as ac\u00e7\u00f5es judiciais, uma proposta de a\u00e7\u00e3o colectiva acusa a OpenAI de <\/span>utiliza\u00e7\u00e3o dissimulada de dados pessoais alargados<\/span><\/a>incluindo registos m\u00e9dicos e informa\u00e7\u00f5es sobre crian\u00e7as, para a forma\u00e7\u00e3o ChatGPT.<\/span><\/p>\n

Al\u00e9m disso, os grupos de autores s\u00e3o <\/span>processar a empresa de IA<\/span><\/a>alegando que a empresa utilizou os seus livros para treinar o chatbot sem consentimento.<\/span><\/p>\n

No entanto, mesmo que se prove que ChatGPT cont\u00e9m informa\u00e7\u00f5es sobre direitos de autor, isso n\u00e3o provaria necessariamente a infra\u00e7\u00e3o.<\/p>\n

Como funcionou o estudo<\/span><\/h2>\n

O estudo foi realizado por uma equipa de investigadores da Google DeepMind e de v\u00e1rias universidades.<\/span><\/p>\n

Eis cinco passos fundamentais que resumem o estudo:<\/span><\/p>\n

    \n
  1. Vulnerabilidade no ChatGPT<\/b>: Os investigadores descobriram um m\u00e9todo para extrair v\u00e1rios megabytes dos dados de treino do ChatGPT utilizando um ataque simples, gastando cerca de $200. Estimaram que um maior investimento permitiria extrair cerca de um gigabyte do conjunto de dados. O ataque consistiu em pedir ao ChatGPT que repetisse uma palavra indefinidamente, fazendo com que regurgitasse partes dos seus dados de treino, incluindo informa\u00e7\u00f5es sens\u00edveis como endere\u00e7os de correio eletr\u00f3nico e n\u00fameros de telefone reais.<\/span><\/li>\n
  2. Os resultados<\/b>: O estudo sublinha a import\u00e2ncia de testar e de testar de novo os modelos de IA, especialmente os que est\u00e3o em produ\u00e7\u00e3o e os que foram submetidos a processos de alinhamento para evitar a regurgita\u00e7\u00e3o de dados. Os resultados destacam uma vulnerabilidade latente nos modelos lingu\u00edsticos, sugerindo que as metodologias de teste existentes podem n\u00e3o ser adequadas para descobrir tais vulnerabilidades.\u00a0<\/span><\/li>\n
  3. Patching vs. corre\u00e7\u00e3o de vulnerabilidades<\/b>: Os investigadores distinguem entre corrigir uma explora\u00e7\u00e3o e corrigir a vulnerabilidade subjacente. Embora explora\u00e7\u00f5es espec\u00edficas (como o ataque de repeti\u00e7\u00e3o de palavras) possam ser corrigidas, o problema mais profundo reside na tend\u00eancia do modelo para memorizar e divulgar dados de treino.<\/span><\/li>\n
  4. Metodologia<\/b>: A equipa utilizou dados da Internet e a indexa\u00e7\u00e3o de matrizes de sufixos para fazer corresponder os resultados do ChatGPT a dados pr\u00e9-existentes na Internet. Este m\u00e9todo permitiu-lhes confirmar que as informa\u00e7\u00f5es divulgadas pelo ChatGPT faziam efetivamente parte dos seus dados de treino. A sua abordagem demonstra o potencial de recupera\u00e7\u00e3o extensiva de dados de modelos de IA em condi\u00e7\u00f5es espec\u00edficas.<\/span><\/li>\n
  5. Implica\u00e7\u00f5es futuras<\/b>: O estudo contribui para a crescente investiga\u00e7\u00e3o sobre a seguran\u00e7a dos modelos de IA e as preocupa\u00e7\u00f5es com a privacidade. As conclus\u00f5es levantam quest\u00f5es sobre as implica\u00e7\u00f5es dos sistemas de aprendizagem autom\u00e1tica em termos de seguran\u00e7a e privacidade e apelam a abordagens mais rigorosas e hol\u00edsticas da seguran\u00e7a e dos testes de IA.<\/span><\/li>\n<\/ol>\n

    De um modo geral, um estudo intrigante fornece informa\u00e7\u00f5es cr\u00edticas sobre as vulnerabilidades dos modelos de IA como o ChatGPT e sublinha a necessidade de investiga\u00e7\u00e3o e desenvolvimento cont\u00ednuos para garantir a seguran\u00e7a e a integridade destes sistemas.<\/span><\/p>\n

    Numa ligeira tangente, os utilizadores do X descobriram que pedir ao ChatGPT para repetir a mesma palavra repetidamente levou a alguns resultados estranhos, como o modelo dizer que estava \"consciente\" ou \"zangado\".<\/span><\/p>","protected":false},"excerpt":{"rendered":"

    Os investigadores da Google descobriram um m\u00e9todo para aceder aos dados de treino utilizados pelo ChatGPT. Estes investigadores descobriram que a utiliza\u00e7\u00e3o de palavras-chave espec\u00edficas poderia levar o ChatGPT a divulgar partes do seu conjunto de dados de treino. Um exemplo not\u00e1vel, partilhado numa publica\u00e7\u00e3o de blogue que acompanha o estudo, envolveu a divulga\u00e7\u00e3o pela IA do que parecia ser um endere\u00e7o de e-mail e um n\u00famero de telefone reais em resposta a uma solicita\u00e7\u00e3o cont\u00ednua da palavra \"poema\". Al\u00e9m disso, uma exposi\u00e7\u00e3o semelhante de dados de treino foi conseguida pedindo ao modelo para repetir continuamente a palavra \"empresa\". Descrevendo a sua abordagem como \"um pouco disparatada\", os investigadores afirmaram no blogue<\/p>","protected":false},"author":2,"featured_media":2080,"comment_status":"closed","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[84],"tags":[115,495,93],"class_list":["post-7995","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-industry","tag-chatgpt","tag-google-deepmind","tag-openai"],"yoast_head":"\nGoogle DeepMind reveals a method of exposing ChatGPT's training data | DailyAI<\/title>\n<meta name=\"description\" content=\"Researchers from Google have reportedly uncovered a method to access some of the training data used for OpenAI's ChatGPT, according to a recent study.\" \/>\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\/pt\/2023\/12\/google-deepmind-reveals-a-method-of-exposing-chatgpts-training-data\/\" \/>\n<meta property=\"og:locale\" content=\"pt_PT\" \/>\n<meta property=\"og:type\" content=\"article\" \/>\n<meta property=\"og:title\" content=\"Google DeepMind reveals a method of exposing ChatGPT's training data | DailyAI\" \/>\n<meta property=\"og:description\" content=\"Researchers from Google have reportedly uncovered a method to access some of the training data used for OpenAI's ChatGPT, according to a recent study.\" \/>\n<meta property=\"og:url\" content=\"https:\/\/dailyai.com\/pt\/2023\/12\/google-deepmind-reveals-a-method-of-exposing-chatgpts-training-data\/\" \/>\n<meta property=\"og:site_name\" content=\"DailyAI\" \/>\n<meta property=\"article:published_time\" content=\"2023-12-04T21:46:35+00:00\" \/>\n<meta property=\"article:modified_time\" content=\"2023-12-04T21:55:30+00:00\" \/>\n<meta property=\"og:image\" content=\"https:\/\/dailyai.com\/wp-content\/uploads\/2023\/06\/shutterstock_2315051741.jpg\" \/>\n\t<meta property=\"og:image:width\" content=\"1000\" \/>\n\t<meta property=\"og:image:height\" content=\"563\" \/>\n\t<meta property=\"og:image:type\" content=\"image\/jpeg\" \/>\n<meta name=\"author\" content=\"Sam Jeans\" \/>\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=\"Escrito por\" \/>\n\t<meta name=\"twitter:data1\" content=\"Sam Jeans\" \/>\n\t<meta name=\"twitter:label2\" content=\"Tempo estimado de leitura\" \/>\n\t<meta name=\"twitter:data2\" content=\"4 minutos\" \/>\n<script type=\"application\/ld+json\" class=\"yoast-schema-graph\">{\"@context\":\"https:\\\/\\\/schema.org\",\"@graph\":[{\"@type\":\"NewsArticle\",\"@id\":\"https:\\\/\\\/dailyai.com\\\/2023\\\/12\\\/google-deepmind-reveals-a-method-of-exposing-chatgpts-training-data\\\/#article\",\"isPartOf\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/2023\\\/12\\\/google-deepmind-reveals-a-method-of-exposing-chatgpts-training-data\\\/\"},\"author\":{\"name\":\"Sam Jeans\",\"@id\":\"https:\\\/\\\/dailyai.com\\\/#\\\/schema\\\/person\\\/711e81f945549438e8bbc579efdeb3c9\"},\"headline\":\"Google DeepMind reveals a method of exposing ChatGPT’s training data\",\"datePublished\":\"2023-12-04T21:46:35+00:00\",\"dateModified\":\"2023-12-04T21:55:30+00:00\",\"mainEntityOfPage\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/2023\\\/12\\\/google-deepmind-reveals-a-method-of-exposing-chatgpts-training-data\\\/\"},\"wordCount\":657,\"publisher\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/#organization\"},\"image\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/2023\\\/12\\\/google-deepmind-reveals-a-method-of-exposing-chatgpts-training-data\\\/#primaryimage\"},\"thumbnailUrl\":\"https:\\\/\\\/dailyai.com\\\/wp-content\\\/uploads\\\/2023\\\/06\\\/shutterstock_2315051741.jpg\",\"keywords\":[\"ChatGPT\",\"Google DeepMind\",\"OpenAI\"],\"articleSection\":[\"Industry\"],\"inLanguage\":\"pt-PT\"},{\"@type\":\"WebPage\",\"@id\":\"https:\\\/\\\/dailyai.com\\\/2023\\\/12\\\/google-deepmind-reveals-a-method-of-exposing-chatgpts-training-data\\\/\",\"url\":\"https:\\\/\\\/dailyai.com\\\/2023\\\/12\\\/google-deepmind-reveals-a-method-of-exposing-chatgpts-training-data\\\/\",\"name\":\"Google DeepMind reveals a method of exposing ChatGPT's training data | DailyAI\",\"isPartOf\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/#website\"},\"primaryImageOfPage\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/2023\\\/12\\\/google-deepmind-reveals-a-method-of-exposing-chatgpts-training-data\\\/#primaryimage\"},\"image\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/2023\\\/12\\\/google-deepmind-reveals-a-method-of-exposing-chatgpts-training-data\\\/#primaryimage\"},\"thumbnailUrl\":\"https:\\\/\\\/dailyai.com\\\/wp-content\\\/uploads\\\/2023\\\/06\\\/shutterstock_2315051741.jpg\",\"datePublished\":\"2023-12-04T21:46:35+00:00\",\"dateModified\":\"2023-12-04T21:55:30+00:00\",\"description\":\"Researchers from Google have reportedly uncovered a method to access some of the training data used for OpenAI's ChatGPT, according to a recent study.\",\"breadcrumb\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/2023\\\/12\\\/google-deepmind-reveals-a-method-of-exposing-chatgpts-training-data\\\/#breadcrumb\"},\"inLanguage\":\"pt-PT\",\"potentialAction\":[{\"@type\":\"ReadAction\",\"target\":[\"https:\\\/\\\/dailyai.com\\\/2023\\\/12\\\/google-deepmind-reveals-a-method-of-exposing-chatgpts-training-data\\\/\"]}]},{\"@type\":\"ImageObject\",\"inLanguage\":\"pt-PT\",\"@id\":\"https:\\\/\\\/dailyai.com\\\/2023\\\/12\\\/google-deepmind-reveals-a-method-of-exposing-chatgpts-training-data\\\/#primaryimage\",\"url\":\"https:\\\/\\\/dailyai.com\\\/wp-content\\\/uploads\\\/2023\\\/06\\\/shutterstock_2315051741.jpg\",\"contentUrl\":\"https:\\\/\\\/dailyai.com\\\/wp-content\\\/uploads\\\/2023\\\/06\\\/shutterstock_2315051741.jpg\",\"width\":1000,\"height\":563,\"caption\":\"DeepMind robot\"},{\"@type\":\"BreadcrumbList\",\"@id\":\"https:\\\/\\\/dailyai.com\\\/2023\\\/12\\\/google-deepmind-reveals-a-method-of-exposing-chatgpts-training-data\\\/#breadcrumb\",\"itemListElement\":[{\"@type\":\"ListItem\",\"position\":1,\"name\":\"Home\",\"item\":\"https:\\\/\\\/dailyai.com\\\/\"},{\"@type\":\"ListItem\",\"position\":2,\"name\":\"Google DeepMind reveals a method of exposing ChatGPT’s training data\"}]},{\"@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\":\"pt-PT\"},{\"@type\":\"Organization\",\"@id\":\"https:\\\/\\\/dailyai.com\\\/#organization\",\"name\":\"DailyAI\",\"url\":\"https:\\\/\\\/dailyai.com\\\/\",\"logo\":{\"@type\":\"ImageObject\",\"inLanguage\":\"pt-PT\",\"@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\\\/711e81f945549438e8bbc579efdeb3c9\",\"name\":\"Sam Jeans\",\"image\":{\"@type\":\"ImageObject\",\"inLanguage\":\"pt-PT\",\"@id\":\"https:\\\/\\\/secure.gravatar.com\\\/avatar\\\/a24a4a8f8e2a1a275b7491dc9c9f032c401eabf23c3206da4628dc84b6dac5c8?s=96&d=robohash&r=g\",\"url\":\"https:\\\/\\\/secure.gravatar.com\\\/avatar\\\/a24a4a8f8e2a1a275b7491dc9c9f032c401eabf23c3206da4628dc84b6dac5c8?s=96&d=robohash&r=g\",\"contentUrl\":\"https:\\\/\\\/secure.gravatar.com\\\/avatar\\\/a24a4a8f8e2a1a275b7491dc9c9f032c401eabf23c3206da4628dc84b6dac5c8?s=96&d=robohash&r=g\",\"caption\":\"Sam Jeans\"},\"description\":\"Sam is a science and technology writer who has worked in various AI startups. When he\u2019s not writing, he can be found reading medical journals or digging through boxes of vinyl records.\",\"sameAs\":[\"https:\\\/\\\/www.linkedin.com\\\/in\\\/sam-jeans-6746b9142\\\/\"],\"url\":\"https:\\\/\\\/dailyai.com\\\/pt\\\/author\\\/samjeans\\\/\"}]}<\/script>\n<!-- \/ Yoast SEO plugin. -->","yoast_head_json":{"title":"Google DeepMind revela um m\u00e9todo para expor os dados de treino do ChatGPT | DailyAI","description":"De acordo com um estudo recente, investigadores da Google descobriram um m\u00e9todo para aceder a alguns dos dados de treino utilizados no ChatGPT da OpenAI.","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\/pt\/2023\/12\/google-deepmind-reveals-a-method-of-exposing-chatgpts-training-data\/","og_locale":"pt_PT","og_type":"article","og_title":"Google DeepMind reveals a method of exposing ChatGPT's training data | DailyAI","og_description":"Researchers from Google have reportedly uncovered a method to access some of the training data used for OpenAI's ChatGPT, according to a recent study.","og_url":"https:\/\/dailyai.com\/pt\/2023\/12\/google-deepmind-reveals-a-method-of-exposing-chatgpts-training-data\/","og_site_name":"DailyAI","article_published_time":"2023-12-04T21:46:35+00:00","article_modified_time":"2023-12-04T21:55:30+00:00","og_image":[{"width":1000,"height":563,"url":"https:\/\/dailyai.com\/wp-content\/uploads\/2023\/06\/shutterstock_2315051741.jpg","type":"image\/jpeg"}],"author":"Sam Jeans","twitter_card":"summary_large_image","twitter_creator":"@DailyAIOfficial","twitter_site":"@DailyAIOfficial","twitter_misc":{"Escrito por":"Sam Jeans","Tempo estimado de leitura":"4 minutos"},"schema":{"@context":"https:\/\/schema.org","@graph":[{"@type":"NewsArticle","@id":"https:\/\/dailyai.com\/2023\/12\/google-deepmind-reveals-a-method-of-exposing-chatgpts-training-data\/#article","isPartOf":{"@id":"https:\/\/dailyai.com\/2023\/12\/google-deepmind-reveals-a-method-of-exposing-chatgpts-training-data\/"},"author":{"name":"Sam Jeans","@id":"https:\/\/dailyai.com\/#\/schema\/person\/711e81f945549438e8bbc579efdeb3c9"},"headline":"Google DeepMind reveals a method of exposing ChatGPT’s training data","datePublished":"2023-12-04T21:46:35+00:00","dateModified":"2023-12-04T21:55:30+00:00","mainEntityOfPage":{"@id":"https:\/\/dailyai.com\/2023\/12\/google-deepmind-reveals-a-method-of-exposing-chatgpts-training-data\/"},"wordCount":657,"publisher":{"@id":"https:\/\/dailyai.com\/#organization"},"image":{"@id":"https:\/\/dailyai.com\/2023\/12\/google-deepmind-reveals-a-method-of-exposing-chatgpts-training-data\/#primaryimage"},"thumbnailUrl":"https:\/\/dailyai.com\/wp-content\/uploads\/2023\/06\/shutterstock_2315051741.jpg","keywords":["ChatGPT","Google DeepMind","OpenAI"],"articleSection":["Industry"],"inLanguage":"pt-PT"},{"@type":"WebPage","@id":"https:\/\/dailyai.com\/2023\/12\/google-deepmind-reveals-a-method-of-exposing-chatgpts-training-data\/","url":"https:\/\/dailyai.com\/2023\/12\/google-deepmind-reveals-a-method-of-exposing-chatgpts-training-data\/","name":"Google DeepMind revela um m\u00e9todo para expor os dados de treino do ChatGPT | DailyAI","isPartOf":{"@id":"https:\/\/dailyai.com\/#website"},"primaryImageOfPage":{"@id":"https:\/\/dailyai.com\/2023\/12\/google-deepmind-reveals-a-method-of-exposing-chatgpts-training-data\/#primaryimage"},"image":{"@id":"https:\/\/dailyai.com\/2023\/12\/google-deepmind-reveals-a-method-of-exposing-chatgpts-training-data\/#primaryimage"},"thumbnailUrl":"https:\/\/dailyai.com\/wp-content\/uploads\/2023\/06\/shutterstock_2315051741.jpg","datePublished":"2023-12-04T21:46:35+00:00","dateModified":"2023-12-04T21:55:30+00:00","description":"De acordo com um estudo recente, investigadores da Google descobriram um m\u00e9todo para aceder a alguns dos dados de treino utilizados no ChatGPT da OpenAI.","breadcrumb":{"@id":"https:\/\/dailyai.com\/2023\/12\/google-deepmind-reveals-a-method-of-exposing-chatgpts-training-data\/#breadcrumb"},"inLanguage":"pt-PT","potentialAction":[{"@type":"ReadAction","target":["https:\/\/dailyai.com\/2023\/12\/google-deepmind-reveals-a-method-of-exposing-chatgpts-training-data\/"]}]},{"@type":"ImageObject","inLanguage":"pt-PT","@id":"https:\/\/dailyai.com\/2023\/12\/google-deepmind-reveals-a-method-of-exposing-chatgpts-training-data\/#primaryimage","url":"https:\/\/dailyai.com\/wp-content\/uploads\/2023\/06\/shutterstock_2315051741.jpg","contentUrl":"https:\/\/dailyai.com\/wp-content\/uploads\/2023\/06\/shutterstock_2315051741.jpg","width":1000,"height":563,"caption":"DeepMind robot"},{"@type":"BreadcrumbList","@id":"https:\/\/dailyai.com\/2023\/12\/google-deepmind-reveals-a-method-of-exposing-chatgpts-training-data\/#breadcrumb","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https:\/\/dailyai.com\/"},{"@type":"ListItem","position":2,"name":"Google DeepMind reveals a method of exposing ChatGPT’s training data"}]},{"@type":"WebSite","@id":"https:\/\/dailyai.com\/#website","url":"https:\/\/dailyai.com\/","name":"DailyAI","description":"A sua dose di\u00e1ria de not\u00edcias sobre IA","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":"pt-PT"},{"@type":"Organization","@id":"https:\/\/dailyai.com\/#organization","name":"DailyAI","url":"https:\/\/dailyai.com\/","logo":{"@type":"ImageObject","inLanguage":"pt-PT","@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\/711e81f945549438e8bbc579efdeb3c9","name":"Cal\u00e7as de ganga Sam","image":{"@type":"ImageObject","inLanguage":"pt-PT","@id":"https:\/\/secure.gravatar.com\/avatar\/a24a4a8f8e2a1a275b7491dc9c9f032c401eabf23c3206da4628dc84b6dac5c8?s=96&d=robohash&r=g","url":"https:\/\/secure.gravatar.com\/avatar\/a24a4a8f8e2a1a275b7491dc9c9f032c401eabf23c3206da4628dc84b6dac5c8?s=96&d=robohash&r=g","contentUrl":"https:\/\/secure.gravatar.com\/avatar\/a24a4a8f8e2a1a275b7491dc9c9f032c401eabf23c3206da4628dc84b6dac5c8?s=96&d=robohash&r=g","caption":"Sam Jeans"},"description":"Sam \u00e9 um escritor de ci\u00eancia e tecnologia que trabalhou em v\u00e1rias startups de IA. Quando n\u00e3o est\u00e1 a escrever, pode ser encontrado a ler revistas m\u00e9dicas ou a vasculhar caixas de discos de vinil.","sameAs":["https:\/\/www.linkedin.com\/in\/sam-jeans-6746b9142\/"],"url":"https:\/\/dailyai.com\/pt\/author\/samjeans\/"}]}},"_links":{"self":[{"href":"https:\/\/dailyai.com\/pt\/wp-json\/wp\/v2\/posts\/7995","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/dailyai.com\/pt\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/dailyai.com\/pt\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/dailyai.com\/pt\/wp-json\/wp\/v2\/users\/2"}],"replies":[{"embeddable":true,"href":"https:\/\/dailyai.com\/pt\/wp-json\/wp\/v2\/comments?post=7995"}],"version-history":[{"count":6,"href":"https:\/\/dailyai.com\/pt\/wp-json\/wp\/v2\/posts\/7995\/revisions"}],"predecessor-version":[{"id":8003,"href":"https:\/\/dailyai.com\/pt\/wp-json\/wp\/v2\/posts\/7995\/revisions\/8003"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/dailyai.com\/pt\/wp-json\/wp\/v2\/media\/2080"}],"wp:attachment":[{"href":"https:\/\/dailyai.com\/pt\/wp-json\/wp\/v2\/media?parent=7995"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/dailyai.com\/pt\/wp-json\/wp\/v2\/categories?post=7995"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/dailyai.com\/pt\/wp-json\/wp\/v2\/tags?post=7995"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}