{"id":6203,"date":"2023-10-07T17:15:19","date_gmt":"2023-10-07T17:15:19","guid":{"rendered":"https:\/\/dailyai.com\/?p=6203"},"modified":"2023-10-07T22:49:19","modified_gmt":"2023-10-07T22:49:19","slug":"can-constitutional-ai-solve-the-issue-of-problematic-ai-behavior","status":"publish","type":"post","link":"https:\/\/dailyai.com\/pt\/2023\/10\/can-constitutional-ai-solve-the-issue-of-problematic-ai-behavior\/","title":{"rendered":"Poder\u00e1 a \"IA constitucional\" resolver a quest\u00e3o do comportamento problem\u00e1tico da IA?"},"content":{"rendered":"

\u00c0 medida que os modelos de IA continuam a integrar-se na nossa vida quotidiana, aumentam as preocupa\u00e7\u00f5es sobre as limita\u00e7\u00f5es e a fiabilidade das suas chamadas \"barreiras de prote\u00e7\u00e3o\".<\/b><\/p>\n

Os modelos de IA omnipresentes, como o GPT-3.5\/4\/4V et al., possuem protec\u00e7\u00f5es e medidas de seguran\u00e7a incorporadas para evitar que produzam resultados il\u00edcitos, n\u00e3o \u00e9ticos ou indesejados. <\/span><\/p>\n

No entanto, estas caracter\u00edsticas de seguran\u00e7a est\u00e3o longe de ser imperme\u00e1veis e os modelos est\u00e3o a provar o seu potencial para se soltarem dos seus guarda-corpos - ou sa\u00edrem dos carris, por assim dizer. <\/span><\/p>\n

Parte do problema \u00e9 que as barreiras de prote\u00e7\u00e3o n\u00e3o est\u00e3o a acompanhar a complexidade e a diversidade dos modelos.\u00a0<\/span><\/p>\n

Nas \u00faltimas semanas, a OpenAI, apoiada pela Microsoft, revelou grandes melhorias no ChatGPT, permitindo-lhe interagir utilizando apenas a voz e responder a quest\u00f5es atrav\u00e9s de imagens e texto. Esta vers\u00e3o multimodal com capacidade de imagem do GPT-4 foi apelidada de \"GPT-4V\".<\/span><\/p>\n

Paralelamente, a Meta anunciou o lan\u00e7amento de um assistente de IA, v\u00e1rios personalidades do chatbot de celebridades<\/a> para os utilizadores do WhatsApp e do Instagram, e uma s\u00e9rie de outras funcionalidades de IA simples, como os autocolantes de IA.\u00a0<\/span><\/p>\n

As pessoas manipularam prontamente o Meta <\/span>Autocolantes de IA para gerar<\/span><\/a> imagens c\u00f3micas e chocantes de desenhos animados, como Karl Marx nu ou Mario com uma espingarda de assalto.\u00a0<\/span><\/p>\n

\u00c0 medida que a corrida \u00e0 comercializa\u00e7\u00e3o da IA se intensifica, as salvaguardas concebidas para controlar os comportamentos da IA - e impedir que gerem conte\u00fados nocivos, desinforma\u00e7\u00e3o ou ajudem em actividades il\u00edcitas - est\u00e3o a revelar-se mais fr\u00e1geis.\u00a0<\/span><\/p>\n

Ser\u00e1 a IA constitucional a resposta?<\/span><\/h2>\n

Para combater esta situa\u00e7\u00e3o, as empresas de desenvolvimento de IA est\u00e3o a esfor\u00e7ar-se por criar \"constitui\u00e7\u00f5es de IA\", um conjunto de princ\u00edpios e valores fundamentais aos quais os modelos de IA devem aderir. A startup Antr\u00f3pica<\/a> foi um dos primeiros a defender a \"IA constitucional\" numa <\/span>Papel 2022<\/span><\/a>.<\/span><\/p>\n

A Google DeepMind tamb\u00e9m estabeleceu regras constitucionais para o seu chatbot Pardal em 2022<\/a> para manter conversas \"\u00fateis, correctas e inofensivas\".\u00a0<\/span><\/p>\n

As constitui\u00e7\u00f5es de IA do Anthropic derivam princ\u00edpios de v\u00e1rias fontes, incluindo a Declara\u00e7\u00e3o dos Direitos Humanos da ONU e os termos de servi\u00e7o da Apple. O modelo est\u00e1 equipado com princ\u00edpios morais fundamentais que impulsionam o comportamento de baixo para cima, em vez de imporem protec\u00e7\u00f5es de cima para baixo.\u00a0<\/span><\/p>\n

Em vez de treinar laboriosamente a IA com in\u00fameros exemplos de certo ou errado fornecidos por seres humanos, esta abordagem incorpora um conjunto de regras ou princ\u00edpios - uma \"constitui\u00e7\u00e3o\" - que a IA cumpre.<\/p>\n

Inicialmente, a IA \u00e9 apresentada a uma situa\u00e7\u00e3o, depois \u00e9-lhe pedido que critique a sua resposta e, por \u00faltimo, que afine o seu comportamento com base na solu\u00e7\u00e3o revista.<\/p>\n

De seguida, o sistema mergulha na fase de aprendizagem por refor\u00e7o. Aqui, avalia a qualidade das suas pr\u00f3prias respostas, distinguindo a melhor. Com o tempo, esta autoavalia\u00e7\u00e3o refina o seu comportamento.<\/p>\n

A diferen\u00e7a \u00e9 que a IA utiliza o seu ciclo de feedback para determinar a recompensa, num m\u00e9todo denominado \"RL from AI Feedback\" (RLAIF). Quando confrontada com pedidos potencialmente prejudiciais ou enganadores, a IA n\u00e3o se limita a esquivar-se ou a recusar. Em vez disso, aborda o assunto de frente, explicando porque \u00e9 que esse pedido pode ser problem\u00e1tico.<\/p>\n

\u00c9 um passo em frente na cria\u00e7\u00e3o de m\u00e1quinas que n\u00e3o s\u00f3 computam, mas tamb\u00e9m \"pensam\" de uma forma estruturada.<\/p>\n

Dario Amodei, diretor executivo e cofundador da Anthropic, sublinhou o desafio de compreender o funcionamento interno dos modelos de IA. Ele sugere que a exist\u00eancia de uma constitui\u00e7\u00e3o tornaria as regras transparentes e expl\u00edcitas, assegurando que todos os utilizadores sabem o que esperar.\u00a0<\/span><\/p>\n

\u00c9 importante salientar que tamb\u00e9m oferece um meio de responsabilizar o modelo caso este n\u00e3o cumpra os princ\u00edpios delineados.<\/span><\/p>\n

Apesar destes esfor\u00e7os, as constitui\u00e7\u00f5es de IA n\u00e3o est\u00e3o isentas de falhas e os modelos de criadores como o Anthropic t\u00eam-se apresentado como vulner\u00e1veis a <\/span>fugas da pris\u00e3o<\/span><\/a> como muitos outros.\u00a0<\/span><\/p>\n

N\u00e3o existem vias universalmente aceites para formar modelos de IA seguros e \u00e9ticos<\/span><\/h2>\n

Historicamente, os modelos de IA t\u00eam sido aperfei\u00e7oados utilizando um m\u00e9todo denominado aprendizagem por refor\u00e7o atrav\u00e9s de feedback humano (RLHF), em que as respostas da IA s\u00e3o classificadas como \"boas\" ou \"m\u00e1s\" por grandes equipas de avaliadores humanos.\u00a0<\/span><\/p>\n

Embora eficaz at\u00e9 certo ponto, este m\u00e9todo tem sido criticado pela sua falta de exatid\u00e3o e especificidade. Para garantir a \u00e9tica e a seguran\u00e7a da IA, as empresas est\u00e3o agora a explorar solu\u00e7\u00f5es alternativas.<\/span><\/p>\n

Por exemplo, a OpenAI adoptou a abordagem \"red-teaming\", contratando peritos de v\u00e1rias disciplinas para testar e identificar os pontos fracos dos seus modelos.<\/span><\/p>\n

O sistema da OpenAI funciona em itera\u00e7\u00f5es: o modelo de IA produz resultados, os revisores humanos avaliam e corrigem esses resultados com base em directrizes espec\u00edficas e o modelo aprende com este feedback. Os dados de forma\u00e7\u00e3o destes revisores s\u00e3o vitais para a calibra\u00e7\u00e3o \u00e9tica do modelo.<\/p>\n

O ChatGPT opta frequentemente por uma resposta conservadora quando confrontado com temas controversos ou sens\u00edveis, evitando por vezes uma resposta direta. Isto contrasta com a IA constitucional, em que o modelo deve elucidar as suas reservas quando confrontado com quest\u00f5es potencialmente prejudiciais, demonstrando ativamente um racioc\u00ednio baseado nas suas regras fundamentais.<\/p>\n

Essencialmente, enquanto o ChatGPT se baseia fortemente no feedback humano para a sua orienta\u00e7\u00e3o \u00e9tica, a IA constitucional utiliza um quadro baseado em regras definidas com mecanismos de auto-revis\u00e3o e uma \u00eanfase no racioc\u00ednio transparente.<\/p>\n

No final, \u00e9 prov\u00e1vel que n\u00e3o exista uma abordagem \u00fanica para o desenvolvimento de IAs \"seguras\" - e alguns, como Elon Musk, criticam a no\u00e7\u00e3o de IA \"acordada\" e higienizada. Estudos comprovaram<\/a> que at\u00e9 mesmo as IAs constitucionais podem ser desbloqueadas, manipulando-as para um comportamento imprevis\u00edvel.\u00a0<\/span><\/p>\n

Rebecca Johnson, investigadora em \u00e9tica da IA na Universidade de Sydney, salientou que os engenheiros de IA e os cientistas inform\u00e1ticos abordam frequentemente os problemas com o objetivo de encontrar solu\u00e7\u00f5es definitivas, que nem sempre t\u00eam em conta as complexidades da natureza humana.\u00a0<\/span><\/p>\n

\"Temos de come\u00e7ar a tratar a IA generativa como extens\u00f5es dos seres humanos, s\u00e3o apenas mais um aspeto da humanidade\", afirmou.\u00a0<\/span><\/p>\n

O controlo global da IA como uma esp\u00e9cie de sistema t\u00e9cnico simples s\u00f3 se tornar\u00e1 <\/span>mais dif\u00edcil \u00e0 medida que evolui<\/span><\/a>E o mesmo se pode dizer dos organismos biol\u00f3gicos como n\u00f3s. <\/span><\/p>\n

A diverg\u00eancia, provocada ou n\u00e3o, \u00e9 talvez inevit\u00e1vel. <\/span><\/p>","protected":false},"excerpt":{"rendered":"

\u00c0 medida que os modelos de IA continuam a ser incorporados no nosso quotidiano, aumentam as preocupa\u00e7\u00f5es sobre as limita\u00e7\u00f5es e a fiabilidade das suas chamadas \"barreiras de prote\u00e7\u00e3o\". Os modelos de IA omnipresentes, como o GPT-3.5\/4\/4V et al., incluem protec\u00e7\u00f5es e medidas de seguran\u00e7a incorporadas para evitar que produzam resultados il\u00edcitos, pouco \u00e9ticos ou indesejados. No entanto, estas caracter\u00edsticas de seguran\u00e7a est\u00e3o longe de ser imperme\u00e1veis, e os modelos est\u00e3o a provar o seu potencial para se soltarem das suas protec\u00e7\u00f5es - ou sa\u00edrem dos carris, por assim dizer. Parte do problema \u00e9 que as barreiras de prote\u00e7\u00e3o n\u00e3o est\u00e3o a acompanhar a complexidade e a diversidade dos modelos. Nas \u00faltimas semanas, a OpenAI, apoiada pela Microsoft, revelou grandes melhorias<\/p>","protected":false},"author":2,"featured_media":6204,"comment_status":"closed","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[84],"tags":[148,422,93],"class_list":["post-6203","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-industry","tag-anthropic","tag-constitutional-ai","tag-openai"],"yoast_head":"\nCan "constitutional AI" solve the issue of problematic AI behavior? | DailyAI<\/title>\n<meta name=\"description\" content=\"As AI models continue to embed themselves in our daily lives, concerns over the limitations and reliability of the so-called "guardrails" are mounting.\" \/>\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\/10\/can-constitutional-ai-solve-the-issue-of-problematic-ai-behavior\/\" \/>\n<meta property=\"og:locale\" content=\"pt_PT\" \/>\n<meta property=\"og:type\" content=\"article\" \/>\n<meta property=\"og:title\" content=\"Can "constitutional AI" solve the issue of problematic AI behavior? | DailyAI\" \/>\n<meta property=\"og:description\" content=\"As AI models continue to embed themselves in our daily lives, concerns over the limitations and reliability of the so-called "guardrails" are mounting.\" \/>\n<meta property=\"og:url\" content=\"https:\/\/dailyai.com\/pt\/2023\/10\/can-constitutional-ai-solve-the-issue-of-problematic-ai-behavior\/\" \/>\n<meta property=\"og:site_name\" content=\"DailyAI\" \/>\n<meta property=\"article:published_time\" content=\"2023-10-07T17:15:19+00:00\" \/>\n<meta property=\"article:modified_time\" content=\"2023-10-07T22:49:19+00:00\" \/>\n<meta property=\"og:image\" content=\"https:\/\/dailyai.com\/wp-content\/uploads\/2023\/10\/shutterstock_2276608417-1.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=\"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\\\/10\\\/can-constitutional-ai-solve-the-issue-of-problematic-ai-behavior\\\/#article\",\"isPartOf\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/2023\\\/10\\\/can-constitutional-ai-solve-the-issue-of-problematic-ai-behavior\\\/\"},\"author\":{\"name\":\"Sam Jeans\",\"@id\":\"https:\\\/\\\/dailyai.com\\\/#\\\/schema\\\/person\\\/711e81f945549438e8bbc579efdeb3c9\"},\"headline\":\"Can “constitutional AI” solve the issue of problematic AI behavior?\",\"datePublished\":\"2023-10-07T17:15:19+00:00\",\"dateModified\":\"2023-10-07T22:49:19+00:00\",\"mainEntityOfPage\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/2023\\\/10\\\/can-constitutional-ai-solve-the-issue-of-problematic-ai-behavior\\\/\"},\"wordCount\":915,\"publisher\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/#organization\"},\"image\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/2023\\\/10\\\/can-constitutional-ai-solve-the-issue-of-problematic-ai-behavior\\\/#primaryimage\"},\"thumbnailUrl\":\"https:\\\/\\\/dailyai.com\\\/wp-content\\\/uploads\\\/2023\\\/10\\\/shutterstock_2276608417-1.jpg\",\"keywords\":[\"Anthropic\",\"Constitutional AI\",\"OpenAI\"],\"articleSection\":[\"Industry\"],\"inLanguage\":\"pt-PT\"},{\"@type\":\"WebPage\",\"@id\":\"https:\\\/\\\/dailyai.com\\\/2023\\\/10\\\/can-constitutional-ai-solve-the-issue-of-problematic-ai-behavior\\\/\",\"url\":\"https:\\\/\\\/dailyai.com\\\/2023\\\/10\\\/can-constitutional-ai-solve-the-issue-of-problematic-ai-behavior\\\/\",\"name\":\"Can \\\"constitutional AI\\\" solve the issue of problematic AI behavior? | DailyAI\",\"isPartOf\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/#website\"},\"primaryImageOfPage\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/2023\\\/10\\\/can-constitutional-ai-solve-the-issue-of-problematic-ai-behavior\\\/#primaryimage\"},\"image\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/2023\\\/10\\\/can-constitutional-ai-solve-the-issue-of-problematic-ai-behavior\\\/#primaryimage\"},\"thumbnailUrl\":\"https:\\\/\\\/dailyai.com\\\/wp-content\\\/uploads\\\/2023\\\/10\\\/shutterstock_2276608417-1.jpg\",\"datePublished\":\"2023-10-07T17:15:19+00:00\",\"dateModified\":\"2023-10-07T22:49:19+00:00\",\"description\":\"As AI models continue to embed themselves in our daily lives, concerns over the limitations and reliability of the so-called \\\"guardrails\\\" are mounting.\",\"breadcrumb\":{\"@id\":\"https:\\\/\\\/dailyai.com\\\/2023\\\/10\\\/can-constitutional-ai-solve-the-issue-of-problematic-ai-behavior\\\/#breadcrumb\"},\"inLanguage\":\"pt-PT\",\"potentialAction\":[{\"@type\":\"ReadAction\",\"target\":[\"https:\\\/\\\/dailyai.com\\\/2023\\\/10\\\/can-constitutional-ai-solve-the-issue-of-problematic-ai-behavior\\\/\"]}]},{\"@type\":\"ImageObject\",\"inLanguage\":\"pt-PT\",\"@id\":\"https:\\\/\\\/dailyai.com\\\/2023\\\/10\\\/can-constitutional-ai-solve-the-issue-of-problematic-ai-behavior\\\/#primaryimage\",\"url\":\"https:\\\/\\\/dailyai.com\\\/wp-content\\\/uploads\\\/2023\\\/10\\\/shutterstock_2276608417-1.jpg\",\"contentUrl\":\"https:\\\/\\\/dailyai.com\\\/wp-content\\\/uploads\\\/2023\\\/10\\\/shutterstock_2276608417-1.jpg\",\"width\":1000,\"height\":667,\"caption\":\"Anthropic AI\"},{\"@type\":\"BreadcrumbList\",\"@id\":\"https:\\\/\\\/dailyai.com\\\/2023\\\/10\\\/can-constitutional-ai-solve-the-issue-of-problematic-ai-behavior\\\/#breadcrumb\",\"itemListElement\":[{\"@type\":\"ListItem\",\"position\":1,\"name\":\"Home\",\"item\":\"https:\\\/\\\/dailyai.com\\\/\"},{\"@type\":\"ListItem\",\"position\":2,\"name\":\"Can “constitutional AI” solve the issue of problematic AI behavior?\"}]},{\"@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":"Poder\u00e1 a \"IA constitucional\" resolver a quest\u00e3o do comportamento problem\u00e1tico da IA? | DailyAI","description":"\u00c0 medida que os modelos de IA continuam a integrar-se na nossa vida quotidiana, aumentam as preocupa\u00e7\u00f5es sobre as limita\u00e7\u00f5es e a fiabilidade das chamadas \"barreiras de prote\u00e7\u00e3o\".","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\/10\/can-constitutional-ai-solve-the-issue-of-problematic-ai-behavior\/","og_locale":"pt_PT","og_type":"article","og_title":"Can \"constitutional AI\" solve the issue of problematic AI behavior? | DailyAI","og_description":"As AI models continue to embed themselves in our daily lives, concerns over the limitations and reliability of the so-called \"guardrails\" are mounting.","og_url":"https:\/\/dailyai.com\/pt\/2023\/10\/can-constitutional-ai-solve-the-issue-of-problematic-ai-behavior\/","og_site_name":"DailyAI","article_published_time":"2023-10-07T17:15:19+00:00","article_modified_time":"2023-10-07T22:49:19+00:00","og_image":[{"width":1000,"height":667,"url":"https:\/\/dailyai.com\/wp-content\/uploads\/2023\/10\/shutterstock_2276608417-1.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\/10\/can-constitutional-ai-solve-the-issue-of-problematic-ai-behavior\/#article","isPartOf":{"@id":"https:\/\/dailyai.com\/2023\/10\/can-constitutional-ai-solve-the-issue-of-problematic-ai-behavior\/"},"author":{"name":"Sam Jeans","@id":"https:\/\/dailyai.com\/#\/schema\/person\/711e81f945549438e8bbc579efdeb3c9"},"headline":"Can “constitutional AI” solve the issue of problematic AI behavior?","datePublished":"2023-10-07T17:15:19+00:00","dateModified":"2023-10-07T22:49:19+00:00","mainEntityOfPage":{"@id":"https:\/\/dailyai.com\/2023\/10\/can-constitutional-ai-solve-the-issue-of-problematic-ai-behavior\/"},"wordCount":915,"publisher":{"@id":"https:\/\/dailyai.com\/#organization"},"image":{"@id":"https:\/\/dailyai.com\/2023\/10\/can-constitutional-ai-solve-the-issue-of-problematic-ai-behavior\/#primaryimage"},"thumbnailUrl":"https:\/\/dailyai.com\/wp-content\/uploads\/2023\/10\/shutterstock_2276608417-1.jpg","keywords":["Anthropic","Constitutional AI","OpenAI"],"articleSection":["Industry"],"inLanguage":"pt-PT"},{"@type":"WebPage","@id":"https:\/\/dailyai.com\/2023\/10\/can-constitutional-ai-solve-the-issue-of-problematic-ai-behavior\/","url":"https:\/\/dailyai.com\/2023\/10\/can-constitutional-ai-solve-the-issue-of-problematic-ai-behavior\/","name":"Poder\u00e1 a \"IA constitucional\" resolver a quest\u00e3o do comportamento problem\u00e1tico da IA? | DailyAI","isPartOf":{"@id":"https:\/\/dailyai.com\/#website"},"primaryImageOfPage":{"@id":"https:\/\/dailyai.com\/2023\/10\/can-constitutional-ai-solve-the-issue-of-problematic-ai-behavior\/#primaryimage"},"image":{"@id":"https:\/\/dailyai.com\/2023\/10\/can-constitutional-ai-solve-the-issue-of-problematic-ai-behavior\/#primaryimage"},"thumbnailUrl":"https:\/\/dailyai.com\/wp-content\/uploads\/2023\/10\/shutterstock_2276608417-1.jpg","datePublished":"2023-10-07T17:15:19+00:00","dateModified":"2023-10-07T22:49:19+00:00","description":"\u00c0 medida que os modelos de IA continuam a integrar-se na nossa vida quotidiana, aumentam as preocupa\u00e7\u00f5es sobre as limita\u00e7\u00f5es e a fiabilidade das chamadas \"barreiras de prote\u00e7\u00e3o\".","breadcrumb":{"@id":"https:\/\/dailyai.com\/2023\/10\/can-constitutional-ai-solve-the-issue-of-problematic-ai-behavior\/#breadcrumb"},"inLanguage":"pt-PT","potentialAction":[{"@type":"ReadAction","target":["https:\/\/dailyai.com\/2023\/10\/can-constitutional-ai-solve-the-issue-of-problematic-ai-behavior\/"]}]},{"@type":"ImageObject","inLanguage":"pt-PT","@id":"https:\/\/dailyai.com\/2023\/10\/can-constitutional-ai-solve-the-issue-of-problematic-ai-behavior\/#primaryimage","url":"https:\/\/dailyai.com\/wp-content\/uploads\/2023\/10\/shutterstock_2276608417-1.jpg","contentUrl":"https:\/\/dailyai.com\/wp-content\/uploads\/2023\/10\/shutterstock_2276608417-1.jpg","width":1000,"height":667,"caption":"Anthropic AI"},{"@type":"BreadcrumbList","@id":"https:\/\/dailyai.com\/2023\/10\/can-constitutional-ai-solve-the-issue-of-problematic-ai-behavior\/#breadcrumb","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https:\/\/dailyai.com\/"},{"@type":"ListItem","position":2,"name":"Can “constitutional AI” solve the issue of problematic AI behavior?"}]},{"@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\/6203","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=6203"}],"version-history":[{"count":12,"href":"https:\/\/dailyai.com\/pt\/wp-json\/wp\/v2\/posts\/6203\/revisions"}],"predecessor-version":[{"id":6216,"href":"https:\/\/dailyai.com\/pt\/wp-json\/wp\/v2\/posts\/6203\/revisions\/6216"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/dailyai.com\/pt\/wp-json\/wp\/v2\/media\/6204"}],"wp:attachment":[{"href":"https:\/\/dailyai.com\/pt\/wp-json\/wp\/v2\/media?parent=6203"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/dailyai.com\/pt\/wp-json\/wp\/v2\/categories?post=6203"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/dailyai.com\/pt\/wp-json\/wp\/v2\/tags?post=6203"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}