{"id":2093,"date":"2025-05-19T14:13:35","date_gmt":"2025-05-19T14:13:35","guid":{"rendered":"https:\/\/elink.cat\/blog\/?p=2093"},"modified":"2025-06-18T14:10:59","modified_gmt":"2025-06-18T14:10:59","slug":"que-pot-fer-i-que-no-pot-fer-encara-la-ia-multimodal","status":"publish","type":"post","link":"https:\/\/elink.cat\/blog\/que-pot-fer-i-que-no-pot-fer-encara-la-ia-multimodal\/","title":{"rendered":"Qu\u00e8 pot fer i qu\u00e8 no pot fer (encara) la IA multimodal?"},"content":{"rendered":"<span class=\"span-reading-time rt-reading-time\" style=\"display: block;\"><span class=\"rt-label rt-prefix\">Temps de lectura: <\/span> <span class=\"rt-time\"> 3<\/span> <span class=\"rt-label rt-postfix\">minuts<\/span><\/span><p class=\"p1\">Si portes temps provant GPT-4o, Gemini o Claude, segurament ja t\u2019ha passat: li passes una foto, li parles, li demanes que et respongui amb veu i et meravella\u2026 per\u00f2 tamb\u00e9 et desespera. Perqu\u00e8 no sempre ent\u00e9n b\u00e9, a vegades respon amb informaci\u00f3 gen\u00e8rica o directament s\u2019inventa el que veu.<\/p>\n<p class=\"p4\"><span class=\"s3\">La IA multimodal ha avan\u00e7at molt\u00edssim, per\u00f2 <\/span>encara hi ha una gran dist\u00e0ncia entre el que promet i el que pot fer de manera fiable cada dia.<span class=\"s3\"> I no passa res, \u00e9s part del proc\u00e9s. Per aix\u00f2 avui volem posar una mica de llum sobre <\/span>qu\u00e8 \u00e9s capa\u00e7 de fer b\u00e9 la IA multimodal i quines coses encara estan \u201ca mig fer\u201d<span class=\"s3\">.<\/span><\/p>\n<h3><b>Qu\u00e8 pot fer b\u00e9 la ia multimodal (avui mateix)<\/b><\/h3>\n<p class=\"p1\">La bona not\u00edcia \u00e9s que hi ha aplicacions que ja funcionen for\u00e7a b\u00e9 i poden aportar valor real, tant a nivell personal com professional. Veiem<\/p>\n<ul>\n<li class=\"p4\"><span class=\"s3\">\u00a0<\/span><b>Llegir i entendre imatges senzilles :\u00a0<\/b><b><\/b>Els models com GPT-4o o Claude poden descriure amb bastant precisi\u00f3 imatges clares: gr\u00e0fics, pantalles, esquemes, objectes, etc. \u00c9s molt \u00fatil per interpretar dades visuals o ajudar en accessibilitat.<\/li>\n<li class=\"p4\"><span class=\"s3\">\u00a0<\/span><b>Mantenir una conversa per veu fluida:\u00a0<\/b><b><\/b>Els nous models s\u00f3n capa\u00e7os de mantenir converses en temps real amb to natural, reconeixent emocions i matisos. GPT-4o, per exemple, sorpr\u00e8n pel seu to hum\u00e0 i la seva capacitat de resposta.<\/li>\n<li class=\"p4\"><span class=\"s3\">\u00a0<\/span><b>Resumir i extreure informaci\u00f3 de documents visuals:\u00a0<\/b><b><\/b>Pots passar-li una captura d\u2019un PowerPoint o un fragment d\u2019un PDF, i fer-li preguntes sobre el contingut. No sempre \u00e9s perfecte, per\u00f2 funciona molt b\u00e9 per contextos coneguts i estructurats.<\/li>\n<li class=\"p4\"><span class=\"s3\">\u00a0<\/span><b>Interpretar dades multimodals de forma integrada:\u00a0<\/b><b><\/b>La for\u00e7a real \u00e9s que pot entendre text, imatge i veu en un mateix context. Pots parlar-li d\u2019una imatge mentre l\u2019est\u00e0 mirant i et respon contextualment. Aix\u00f2 \u00e9s nou, i \u00e9s molt potent.<\/li>\n<\/ul>\n<h3><b>Qu\u00e8 encara no pot fer (del tot b\u00e9)<\/b><\/h3>\n<p class=\"p1\">Aqu\u00ed \u00e9s on cal una mica de paci\u00e8ncia i realisme. Aquestes funcionalitats encara tenen moltes limitacions:<\/p>\n<ul>\n<li class=\"p4\"><span class=\"s3\">\u00a0<\/span><b>Entendre imatges complexes o amb molt soroll visual: <\/b><b><\/b>Escenes amb molts elements, text petit o contextos abstractes (com una foto d\u2019una classe amb molts estudiants o un mapa complex) poden confondre el model o portar-lo a donar respostes vagues o incorrectes.<\/li>\n<li class=\"p4\"><span class=\"s3\">\u00a0<\/span><b>Raonar amb imatges i dades combinades de forma precisa :\u00a0<\/b><b><\/b>Si li passes una taula amb n\u00fameros i li demanes una an\u00e0lisi detallada, pot fallar. El raonament matem\u00e0tic o estad\u00edstic encara no \u00e9s consistent, i les respostes poden ser poc fiables.<\/li>\n<li class=\"p4\"><span class=\"s3\">\u00a0<\/span><b>Interaccions multimodals en temps real 100% flu\u00efdes :<\/b><b><\/b>Tot i que es parla molt del \u201ctemps real\u201d, la realitat \u00e9s que encara hi ha lat\u00e8ncies, talls, i respostes que triguen. La flu\u00efdesa total (com si fos una conversa humana amb visualitzaci\u00f3 constant) encara no hi \u00e9s.<\/li>\n<li class=\"p4\"><span class=\"s3\">\u00a0<\/span><b>Respostes completament veraces i precises :<\/b><b><\/b>Com qualsevol model generatiu, pot inventar dades (\u201challucinations\u201d) o interpretar malament el que veu o escolta. Sobretot quan les preguntes s\u00f3n obertes o ambigus.<\/li>\n<\/ul>\n<h3><b>El risc de confondre potencial amb realitat<\/b><\/h3>\n<p class=\"p1\">El gran repte d\u2019aquesta etapa \u00e9s que <span class=\"s1\"><b>els v\u00eddeos promocionals s\u00f3n molt millors que l\u2019experi\u00e8ncia real.<\/b><\/span> I aix\u00f2 pot portar a frustracions, especialment en entorns professionals que esperen una resposta fiable cada vegada.<\/p>\n<p class=\"p1\">Per\u00f2 aix\u00f2 no vol dir que no siguin \u00fatils. Vol dir que cal <span class=\"s1\"><b>entendre molt b\u00e9 el context, els l\u00edmits i els usos adequats<\/b><\/span>. Per exemple, un assistent multimodal \u00e9s ideal per ajudar a navegar una web complexa o entendre una gr\u00e0fica, per\u00f2 no \u00e9s bona idea fer-lo servir per prendre decisions cr\u00edtiques sense supervisi\u00f3.<\/p>\n<h3><b>Cap on evoluciona tot aix\u00f2?<\/b><\/h3>\n<p class=\"p1\">Els pr\u00f2xims mesos veurem millores r\u00e0pides en:<\/p>\n<ul>\n<li class=\"p1\"><span class=\"s1\"><b>Exactitud visual:<\/b><\/span> millor reconeixement de detalls i context d\u2019imatge<\/li>\n<li class=\"p1\"><span class=\"s1\"><b>Control de la veu:<\/b><\/span> to, pauses, emoci\u00f3 m\u00e9s realistes i adaptatius<\/li>\n<li class=\"p1\"><span class=\"s1\"><b>Temps de resposta:<\/b><\/span> converses m\u00e9s flu\u00efdes i menys temps d\u2019espera<\/li>\n<li class=\"p1\"><span class=\"s1\"><b>Integraci\u00f3 amb aplicacions reals:<\/b><\/span> podran actuar sobre sistemes, no nom\u00e9s parlar<\/li>\n<\/ul>\n<p class=\"p1\">I, a mitj\u00e0 termini, veurem <span class=\"s1\"><b>agents multimodals aut\u00f2noms<\/b><\/span> capa\u00e7os no nom\u00e9s d\u2019interpretar informaci\u00f3 diversa, sin\u00f3 de fer accions concretes dins entorns empresarials, operatius o creatius.<\/p>\n<p class=\"p1\">La multimodalitat en IA no \u00e9s ci\u00e8ncia ficci\u00f3, per\u00f2 tampoc \u00e9s m\u00e0gia. \u00c9s una realitat amb molt potencial que <span class=\"s1\"><b>ja comen\u00e7a a ser \u00fatil, per\u00f2 que encara t\u00e9 molt recorregut per fer<\/b><\/span>.<\/p>\n<p class=\"p1\">Si saps qu\u00e8 li pots demanar (i qu\u00e8 no), pots comen\u00e7ar a aprofitar-la ara mateix. Per\u00f2 si vols que et resolgui la vida com a \u201csiri del futur\u201d, encara haur\u00e0s d\u2019esperar una mica.<\/p>\n<p class=\"p1\">Al final, la clau est\u00e0 en l\u2019equilibri: aprofitar el que funciona, detectar el que falla, i continuar explorant com aquesta nova forma d\u2019interacci\u00f3 amb la tecnologia ens pot ajudar a crear productes i serveis molt m\u00e9s humans.<\/p>\n","protected":false},"excerpt":{"rendered":"<p><span class=\"span-reading-time rt-reading-time\" style=\"display: block;\"><span class=\"rt-label rt-prefix\">Temps de lectura: <\/span> <span class=\"rt-time\"> 3<\/span> <span class=\"rt-label rt-postfix\">minuts<\/span><\/span>Si portes temps provant GPT-4o, Gemini o Claude, segurament ja t\u2019ha passat: li passes una foto, li parles, li demanes que et respongui amb veu i et meravella\u2026 per\u00f2 tamb\u00e9<\/p>\n","protected":false},"author":1,"featured_media":2100,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"_crdt_document":"","breadcrumbs_single_post":"","page_title_panel":"","breadcrumbs_single_page":"","single_page_alignment":"","single_page_margin":"","page_structure_type":"","content_style_source":"","content_style":"","blog_post_streched_ed":"","blog_page_streched_ed":"","has_transparent_header":"","disable_transparent_header":"","vertical_spacing_source":"","content_area_spacing":"","single_post_content_background":"","single_page_content_background":"","single_post_boxed_content_spacing":"","single_page_boxed_content_spacing":"","single_post_content_boxed_radius":"","single_page_content_boxed_radius":"","disable_featured_image":"","disable_post_tags":"","disable_author_box":"","disable_posts_navigation":"","disable_comments":"","disable_related_posts":"","disable_header":"","disable_footer":"","footnotes":""},"categories":[29],"tags":[],"class_list":["post-2093","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-futur","rishi-post"],"yoast_head":"<!-- This site is optimized with the Yoast SEO plugin v27.3 - https:\/\/yoast.com\/product\/yoast-seo-wordpress\/ -->\n<title>Qu\u00e8 pot fer i qu\u00e8 no pot fer (encara) la IA multimodal? - Blog Elinkcat<\/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:\/\/elink.cat\/blog\/que-pot-fer-i-que-no-pot-fer-encara-la-ia-multimodal\/\" \/>\n<meta property=\"og:locale\" content=\"ca_ES\" \/>\n<meta property=\"og:type\" content=\"article\" \/>\n<meta property=\"og:title\" content=\"Qu\u00e8 pot fer i qu\u00e8 no pot fer (encara) la IA multimodal? - Blog Elinkcat\" \/>\n<meta property=\"og:description\" content=\"Temps de lectura:  3 minutsSi portes temps provant GPT-4o, Gemini o Claude, segurament ja t\u2019ha passat: li passes una foto, li parles, li demanes que et respongui amb veu i et meravella\u2026 per\u00f2 tamb\u00e9\" \/>\n<meta property=\"og:url\" content=\"https:\/\/elink.cat\/blog\/que-pot-fer-i-que-no-pot-fer-encara-la-ia-multimodal\/\" \/>\n<meta property=\"og:site_name\" content=\"Blog Elinkcat\" \/>\n<meta property=\"article:published_time\" content=\"2025-05-19T14:13:35+00:00\" \/>\n<meta property=\"article:modified_time\" content=\"2025-06-18T14:10:59+00:00\" \/>\n<meta property=\"og:image\" content=\"https:\/\/elink.cat\/blog\/wp-content\/uploads\/2025\/05\/fer-no-fer-IA-multimodal.jpeg\" \/>\n\t<meta property=\"og:image:width\" content=\"1024\" \/>\n\t<meta property=\"og:image:height\" content=\"1024\" \/>\n\t<meta property=\"og:image:type\" content=\"image\/jpeg\" \/>\n<meta name=\"author\" content=\"\u00d2scar Junyent\" \/>\n<meta name=\"twitter:card\" content=\"summary_large_image\" \/>\n<meta name=\"twitter:label1\" content=\"Written by\" \/>\n\t<meta name=\"twitter:data1\" content=\"\u00d2scar Junyent\" \/>\n\t<meta name=\"twitter:label2\" content=\"Est. reading time\" \/>\n\t<meta name=\"twitter:data2\" content=\"4 minuts\" \/>\n<script type=\"application\/ld+json\" class=\"yoast-schema-graph\">{\"@context\":\"https:\\\/\\\/schema.org\",\"@graph\":[{\"@type\":\"Article\",\"@id\":\"https:\\\/\\\/elink.cat\\\/blog\\\/que-pot-fer-i-que-no-pot-fer-encara-la-ia-multimodal\\\/#article\",\"isPartOf\":{\"@id\":\"https:\\\/\\\/elink.cat\\\/blog\\\/que-pot-fer-i-que-no-pot-fer-encara-la-ia-multimodal\\\/\"},\"author\":{\"name\":\"\u00d2scar Junyent\",\"@id\":\"https:\\\/\\\/elink.cat\\\/blog\\\/#\\\/schema\\\/person\\\/13577ee4b0279d498b46e86c8798afe2\"},\"headline\":\"Qu\u00e8 pot fer i qu\u00e8 no pot fer (encara) la IA multimodal?\",\"datePublished\":\"2025-05-19T14:13:35+00:00\",\"dateModified\":\"2025-06-18T14:10:59+00:00\",\"mainEntityOfPage\":{\"@id\":\"https:\\\/\\\/elink.cat\\\/blog\\\/que-pot-fer-i-que-no-pot-fer-encara-la-ia-multimodal\\\/\"},\"wordCount\":819,\"publisher\":{\"@id\":\"https:\\\/\\\/elink.cat\\\/blog\\\/#organization\"},\"image\":{\"@id\":\"https:\\\/\\\/elink.cat\\\/blog\\\/que-pot-fer-i-que-no-pot-fer-encara-la-ia-multimodal\\\/#primaryimage\"},\"thumbnailUrl\":\"https:\\\/\\\/elink.cat\\\/blog\\\/wp-content\\\/uploads\\\/2025\\\/05\\\/fer-no-fer-IA-multimodal.jpeg\",\"articleSection\":[\"Futur de la interacci\u00f3 amb la IA\"],\"inLanguage\":\"ca\"},{\"@type\":\"WebPage\",\"@id\":\"https:\\\/\\\/elink.cat\\\/blog\\\/que-pot-fer-i-que-no-pot-fer-encara-la-ia-multimodal\\\/\",\"url\":\"https:\\\/\\\/elink.cat\\\/blog\\\/que-pot-fer-i-que-no-pot-fer-encara-la-ia-multimodal\\\/\",\"name\":\"Qu\u00e8 pot fer i qu\u00e8 no pot fer (encara) la IA multimodal? - Blog Elinkcat\",\"isPartOf\":{\"@id\":\"https:\\\/\\\/elink.cat\\\/blog\\\/#website\"},\"primaryImageOfPage\":{\"@id\":\"https:\\\/\\\/elink.cat\\\/blog\\\/que-pot-fer-i-que-no-pot-fer-encara-la-ia-multimodal\\\/#primaryimage\"},\"image\":{\"@id\":\"https:\\\/\\\/elink.cat\\\/blog\\\/que-pot-fer-i-que-no-pot-fer-encara-la-ia-multimodal\\\/#primaryimage\"},\"thumbnailUrl\":\"https:\\\/\\\/elink.cat\\\/blog\\\/wp-content\\\/uploads\\\/2025\\\/05\\\/fer-no-fer-IA-multimodal.jpeg\",\"datePublished\":\"2025-05-19T14:13:35+00:00\",\"dateModified\":\"2025-06-18T14:10:59+00:00\",\"breadcrumb\":{\"@id\":\"https:\\\/\\\/elink.cat\\\/blog\\\/que-pot-fer-i-que-no-pot-fer-encara-la-ia-multimodal\\\/#breadcrumb\"},\"inLanguage\":\"ca\",\"potentialAction\":[{\"@type\":\"ReadAction\",\"target\":[\"https:\\\/\\\/elink.cat\\\/blog\\\/que-pot-fer-i-que-no-pot-fer-encara-la-ia-multimodal\\\/\"]}]},{\"@type\":\"ImageObject\",\"inLanguage\":\"ca\",\"@id\":\"https:\\\/\\\/elink.cat\\\/blog\\\/que-pot-fer-i-que-no-pot-fer-encara-la-ia-multimodal\\\/#primaryimage\",\"url\":\"https:\\\/\\\/elink.cat\\\/blog\\\/wp-content\\\/uploads\\\/2025\\\/05\\\/fer-no-fer-IA-multimodal.jpeg\",\"contentUrl\":\"https:\\\/\\\/elink.cat\\\/blog\\\/wp-content\\\/uploads\\\/2025\\\/05\\\/fer-no-fer-IA-multimodal.jpeg\",\"width\":1024,\"height\":1024},{\"@type\":\"BreadcrumbList\",\"@id\":\"https:\\\/\\\/elink.cat\\\/blog\\\/que-pot-fer-i-que-no-pot-fer-encara-la-ia-multimodal\\\/#breadcrumb\",\"itemListElement\":[{\"@type\":\"ListItem\",\"position\":1,\"name\":\"Home\",\"item\":\"https:\\\/\\\/elink.cat\\\/blog\\\/\"},{\"@type\":\"ListItem\",\"position\":2,\"name\":\"Qu\u00e8 pot fer i qu\u00e8 no pot fer (encara) la IA multimodal?\"}]},{\"@type\":\"WebSite\",\"@id\":\"https:\\\/\\\/elink.cat\\\/blog\\\/#website\",\"url\":\"https:\\\/\\\/elink.cat\\\/blog\\\/\",\"name\":\"Blog Elinkcat\",\"description\":\"\",\"publisher\":{\"@id\":\"https:\\\/\\\/elink.cat\\\/blog\\\/#organization\"},\"potentialAction\":[{\"@type\":\"SearchAction\",\"target\":{\"@type\":\"EntryPoint\",\"urlTemplate\":\"https:\\\/\\\/elink.cat\\\/blog\\\/?s={search_term_string}\"},\"query-input\":{\"@type\":\"PropertyValueSpecification\",\"valueRequired\":true,\"valueName\":\"search_term_string\"}}],\"inLanguage\":\"ca\"},{\"@type\":\"Organization\",\"@id\":\"https:\\\/\\\/elink.cat\\\/blog\\\/#organization\",\"name\":\"Blog Elinkcat\",\"url\":\"https:\\\/\\\/elink.cat\\\/blog\\\/\",\"logo\":{\"@type\":\"ImageObject\",\"inLanguage\":\"ca\",\"@id\":\"https:\\\/\\\/elink.cat\\\/blog\\\/#\\\/schema\\\/logo\\\/image\\\/\",\"url\":\"https:\\\/\\\/elink.cat\\\/blog\\\/wp-content\\\/uploads\\\/2024\\\/01\\\/cropped-elinkcat-default-light-bg.png\",\"contentUrl\":\"https:\\\/\\\/elink.cat\\\/blog\\\/wp-content\\\/uploads\\\/2024\\\/01\\\/cropped-elinkcat-default-light-bg.png\",\"width\":1278,\"height\":127,\"caption\":\"Blog Elinkcat\"},\"image\":{\"@id\":\"https:\\\/\\\/elink.cat\\\/blog\\\/#\\\/schema\\\/logo\\\/image\\\/\"}},{\"@type\":\"Person\",\"@id\":\"https:\\\/\\\/elink.cat\\\/blog\\\/#\\\/schema\\\/person\\\/13577ee4b0279d498b46e86c8798afe2\",\"name\":\"\u00d2scar Junyent\",\"sameAs\":[\"https:\\\/\\\/elink.cat\\\/blog\"],\"url\":\"https:\\\/\\\/elink.cat\\\/blog\\\/author\\\/ojunyentelink-cat\\\/\"}]}<\/script>\n<!-- \/ Yoast SEO plugin. -->","yoast_head_json":{"title":"Qu\u00e8 pot fer i qu\u00e8 no pot fer (encara) la IA multimodal? - Blog Elinkcat","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:\/\/elink.cat\/blog\/que-pot-fer-i-que-no-pot-fer-encara-la-ia-multimodal\/","og_locale":"ca_ES","og_type":"article","og_title":"Qu\u00e8 pot fer i qu\u00e8 no pot fer (encara) la IA multimodal? - Blog Elinkcat","og_description":"Temps de lectura:  3 minutsSi portes temps provant GPT-4o, Gemini o Claude, segurament ja t\u2019ha passat: li passes una foto, li parles, li demanes que et respongui amb veu i et meravella\u2026 per\u00f2 tamb\u00e9","og_url":"https:\/\/elink.cat\/blog\/que-pot-fer-i-que-no-pot-fer-encara-la-ia-multimodal\/","og_site_name":"Blog Elinkcat","article_published_time":"2025-05-19T14:13:35+00:00","article_modified_time":"2025-06-18T14:10:59+00:00","og_image":[{"width":1024,"height":1024,"url":"https:\/\/elink.cat\/blog\/wp-content\/uploads\/2025\/05\/fer-no-fer-IA-multimodal.jpeg","type":"image\/jpeg"}],"author":"\u00d2scar Junyent","twitter_card":"summary_large_image","twitter_misc":{"Written by":"\u00d2scar Junyent","Est. reading time":"4 minuts"},"schema":{"@context":"https:\/\/schema.org","@graph":[{"@type":"Article","@id":"https:\/\/elink.cat\/blog\/que-pot-fer-i-que-no-pot-fer-encara-la-ia-multimodal\/#article","isPartOf":{"@id":"https:\/\/elink.cat\/blog\/que-pot-fer-i-que-no-pot-fer-encara-la-ia-multimodal\/"},"author":{"name":"\u00d2scar Junyent","@id":"https:\/\/elink.cat\/blog\/#\/schema\/person\/13577ee4b0279d498b46e86c8798afe2"},"headline":"Qu\u00e8 pot fer i qu\u00e8 no pot fer (encara) la IA multimodal?","datePublished":"2025-05-19T14:13:35+00:00","dateModified":"2025-06-18T14:10:59+00:00","mainEntityOfPage":{"@id":"https:\/\/elink.cat\/blog\/que-pot-fer-i-que-no-pot-fer-encara-la-ia-multimodal\/"},"wordCount":819,"publisher":{"@id":"https:\/\/elink.cat\/blog\/#organization"},"image":{"@id":"https:\/\/elink.cat\/blog\/que-pot-fer-i-que-no-pot-fer-encara-la-ia-multimodal\/#primaryimage"},"thumbnailUrl":"https:\/\/elink.cat\/blog\/wp-content\/uploads\/2025\/05\/fer-no-fer-IA-multimodal.jpeg","articleSection":["Futur de la interacci\u00f3 amb la IA"],"inLanguage":"ca"},{"@type":"WebPage","@id":"https:\/\/elink.cat\/blog\/que-pot-fer-i-que-no-pot-fer-encara-la-ia-multimodal\/","url":"https:\/\/elink.cat\/blog\/que-pot-fer-i-que-no-pot-fer-encara-la-ia-multimodal\/","name":"Qu\u00e8 pot fer i qu\u00e8 no pot fer (encara) la IA multimodal? - Blog Elinkcat","isPartOf":{"@id":"https:\/\/elink.cat\/blog\/#website"},"primaryImageOfPage":{"@id":"https:\/\/elink.cat\/blog\/que-pot-fer-i-que-no-pot-fer-encara-la-ia-multimodal\/#primaryimage"},"image":{"@id":"https:\/\/elink.cat\/blog\/que-pot-fer-i-que-no-pot-fer-encara-la-ia-multimodal\/#primaryimage"},"thumbnailUrl":"https:\/\/elink.cat\/blog\/wp-content\/uploads\/2025\/05\/fer-no-fer-IA-multimodal.jpeg","datePublished":"2025-05-19T14:13:35+00:00","dateModified":"2025-06-18T14:10:59+00:00","breadcrumb":{"@id":"https:\/\/elink.cat\/blog\/que-pot-fer-i-que-no-pot-fer-encara-la-ia-multimodal\/#breadcrumb"},"inLanguage":"ca","potentialAction":[{"@type":"ReadAction","target":["https:\/\/elink.cat\/blog\/que-pot-fer-i-que-no-pot-fer-encara-la-ia-multimodal\/"]}]},{"@type":"ImageObject","inLanguage":"ca","@id":"https:\/\/elink.cat\/blog\/que-pot-fer-i-que-no-pot-fer-encara-la-ia-multimodal\/#primaryimage","url":"https:\/\/elink.cat\/blog\/wp-content\/uploads\/2025\/05\/fer-no-fer-IA-multimodal.jpeg","contentUrl":"https:\/\/elink.cat\/blog\/wp-content\/uploads\/2025\/05\/fer-no-fer-IA-multimodal.jpeg","width":1024,"height":1024},{"@type":"BreadcrumbList","@id":"https:\/\/elink.cat\/blog\/que-pot-fer-i-que-no-pot-fer-encara-la-ia-multimodal\/#breadcrumb","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https:\/\/elink.cat\/blog\/"},{"@type":"ListItem","position":2,"name":"Qu\u00e8 pot fer i qu\u00e8 no pot fer (encara) la IA multimodal?"}]},{"@type":"WebSite","@id":"https:\/\/elink.cat\/blog\/#website","url":"https:\/\/elink.cat\/blog\/","name":"Blog Elinkcat","description":"","publisher":{"@id":"https:\/\/elink.cat\/blog\/#organization"},"potentialAction":[{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https:\/\/elink.cat\/blog\/?s={search_term_string}"},"query-input":{"@type":"PropertyValueSpecification","valueRequired":true,"valueName":"search_term_string"}}],"inLanguage":"ca"},{"@type":"Organization","@id":"https:\/\/elink.cat\/blog\/#organization","name":"Blog Elinkcat","url":"https:\/\/elink.cat\/blog\/","logo":{"@type":"ImageObject","inLanguage":"ca","@id":"https:\/\/elink.cat\/blog\/#\/schema\/logo\/image\/","url":"https:\/\/elink.cat\/blog\/wp-content\/uploads\/2024\/01\/cropped-elinkcat-default-light-bg.png","contentUrl":"https:\/\/elink.cat\/blog\/wp-content\/uploads\/2024\/01\/cropped-elinkcat-default-light-bg.png","width":1278,"height":127,"caption":"Blog Elinkcat"},"image":{"@id":"https:\/\/elink.cat\/blog\/#\/schema\/logo\/image\/"}},{"@type":"Person","@id":"https:\/\/elink.cat\/blog\/#\/schema\/person\/13577ee4b0279d498b46e86c8798afe2","name":"\u00d2scar Junyent","sameAs":["https:\/\/elink.cat\/blog"],"url":"https:\/\/elink.cat\/blog\/author\/ojunyentelink-cat\/"}]}},"_links":{"self":[{"href":"https:\/\/elink.cat\/blog\/wp-json\/wp\/v2\/posts\/2093","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/elink.cat\/blog\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/elink.cat\/blog\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/elink.cat\/blog\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/elink.cat\/blog\/wp-json\/wp\/v2\/comments?post=2093"}],"version-history":[{"count":7,"href":"https:\/\/elink.cat\/blog\/wp-json\/wp\/v2\/posts\/2093\/revisions"}],"predecessor-version":[{"id":2101,"href":"https:\/\/elink.cat\/blog\/wp-json\/wp\/v2\/posts\/2093\/revisions\/2101"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/elink.cat\/blog\/wp-json\/wp\/v2\/media\/2100"}],"wp:attachment":[{"href":"https:\/\/elink.cat\/blog\/wp-json\/wp\/v2\/media?parent=2093"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/elink.cat\/blog\/wp-json\/wp\/v2\/categories?post=2093"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/elink.cat\/blog\/wp-json\/wp\/v2\/tags?post=2093"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}