{"id":22651,"date":"2026-09-23T18:51:37","date_gmt":"2026-09-23T17:51:37","guid":{"rendered":"https:\/\/loxias.ai\/?p=22651"},"modified":"2026-09-29T15:23:15","modified_gmt":"2026-09-29T14:23:15","slug":"mcp-how-ai-starts-talking-to-your-companys-data","status":"publish","type":"post","link":"https:\/\/loxias.ai\/es\/blog\/mcp-how-ai-starts-talking-to-your-companys-data\/","title":{"rendered":"MCP: C\u00f3mo el IA comienza a hablar con los datos de tu empresa"},"content":{"rendered":"\n<p class=\"wp-block-paragraph\"><em>Aprende c\u00f3mo Model Context Protocol, o MCP, conecta la inteligencia artificial generativa con los datos empresariales, mejora el acceso a las ideas comerciales y presenta nuevas consideraciones de gobernanza y seguridad.<\/em><\/p>\n\n\n\n<p class=\"wp-block-paragraph\">Un est\u00e1ndar abierto dise\u00f1ado para conectar la inteligencia artificial generativa con los sistemas empresariales est\u00e1 cambiando qui\u00e9n dentro de las empresas puede hacer preguntas sobre sus datos y con qu\u00e9 rapidez pueden obtener respuestas.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">Es lunes a las 9 a.m. El director de marketing quiere entender si el aumento de quejas sobre la aplicaci\u00f3n de la empresa en las redes sociales coincide con una disminuci\u00f3n en las calificaciones del servicio al cliente. La plataforma de monitoreo social muestra una parte de la historia, mientras que el sistema de servicio al cliente muestra otra. Conectar los dos usualmente significa enviar una solicitud al equipo de datos, esperar en una cola y recibir un informe d\u00edas despu\u00e9s. A veces, para cuando llega, la reuni\u00f3n urgente que necesitaba la informaci\u00f3n ya ha sucedido.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">That gap between a question and an answer is one of the less visible costs of data driven management. And it is precisely the kind of gap that a technical standard still relatively unknown outside technology teams can help reduce: MCP, short for <em><a href=\"https:\/\/loxias.ai\/es\/blog\/mcp-the-protocol-thats-reshaping-ai-to-the-world\/\" data-type=\"link\" data-id=\"https:\/\/loxias.ai\/blog\/mcp-the-protocol-thats-reshaping-ai-to-the-world\/\">Model Context Protocol<\/a><\/em>.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">Una empresa instala un servidor MCP que expone datos y acciones del sistema de una manera estandarizada. Un asistente de inteligencia artificial se conecta a ese servidor y, cuando se conecta a varios servidores al mismo tiempo, puede consultar diferentes fuentes y combinar la informaci\u00f3n dentro de la conversaci\u00f3n en s\u00ed.<\/p>\n\n\n\n<div style=\"height:25px\" aria-hidden=\"true\" class=\"wp-block-spacer\"><\/div>\n\n\n\n<h2 class=\"wp-block-heading\">Un est\u00e1ndar com\u00fan para conectar la inteligencia artificial generativa a los sistemas empresariales<\/h2>\n\n\n\n<p class=\"wp-block-paragraph\">Launched by <a href=\"https:\/\/www.anthropic.com\/news\/model-context-protocol\" data-type=\"link\" data-id=\"https:\/\/www.anthropic.com\/news\/model-context-protocol\" rel=\"nofollow noopener\" target=\"_blank\">Anthropic <\/a>in November 2024, MCP is an open protocol that defines a standardized way to connect AI models to external systems such as databases, marketing platforms, and internal business applications. Before MCP, each integration was essentially a custom project.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">Anthropic itself compara MCP con USB C. En el pasado, cada dispositivo ten\u00eda su propio cargador. Hoy en d\u00eda, un solo cable puede funcionar en tel\u00e9fonos, port\u00e1tiles y auriculares.<\/p>\n\n\n\n<blockquote class=\"wp-block-quote is-layout-flow wp-block-quote-is-layout-flow\">\n<p class=\"wp-block-paragraph\">T<em>here is an even more familiar comparison:<\/em> <em><a href=\"https:\/\/www.bcb.gov.br\/en\/financialstability\/pix_en\" target=\"_blank\" rel=\"noopener nofollow\">digital wallets<\/a>. Before digital wallets, transferring money between banks involved different rules, processing times, and fees. With a common standard, any application could communicate with any participating bank almost instantly. MCP does something similar for AI, with one important distinction. The digital wallets has a regulator that mandates and certifies participation. MCP, by contrast, is an open specification adopted voluntarily, and each tool may implement different parts of it using its own authentication models. The benefit, therefore, is not the elimination of integration work, but a significant reduction in duplicated effort.<\/em><\/p>\n<\/blockquote>\n\n\n\n<p class=\"wp-block-paragraph\">According to Anthropic, adoption has been rapid. By December 2025, the ecosystem already included more than 10,000 active public MCP servers, and the protocol had been adopted by products including <em>ChatGPT, Cursor, Gemini, Microsoft Copilot, and Visual Studio Code.<\/em><\/p>\n\n\n\n<p class=\"wp-block-paragraph\">In December of that year, Anthropic donated <strong>MCP to the Agentic AI Foundation<\/strong>, an initiative associated with the Linux Foundation and co founded by<em> Anthropic, Block, and OpenAI<\/em>, with support from Google, Microsoft, AWS, Cloudflare, and Bloomberg.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">Para los responsables de decisiones tecnol\u00f3gicas, ese cambio es importante. El est\u00e1ndar ya no est\u00e1 controlado por un solo proveedor.<\/p>\n\n\n\n<div style=\"height:25px\" aria-hidden=\"true\" class=\"wp-block-spacer\"><\/div>\n\n\n\n<h2 class=\"wp-block-heading\">\u00bfQu\u00e9 cambios MCP para empresas<\/h2>\n\n\n\n<p class=\"wp-block-paragraph\">El primer impacto es una mayor autonom\u00eda.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">Con los servidores MCP que exponen datos de una plataforma de escucha social y un sistema de servicio al cliente, el director de marketing de nuestro ejemplo puede preguntar directamente al asistente de IA si el aumento de quejas sobre la aplicaci\u00f3n coincide con un declive en las calificaciones del servicio al cliente.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">El IA puede consultar ambas fuentes, combinar la informaci\u00f3n y proporcionar una respuesta en lenguaje natural. El director puede solicitar un gr\u00e1fico o un resumen ejecutivo para el liderazgo senior dentro de la misma conversaci\u00f3n.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">El segundo impacto es una mayor eficiencia para los equipos t\u00e9cnicos.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">Los equipos de datos pueden dedicar menos tiempo a responder a solicitudes de informes repetitivos y m\u00e1s tiempo a construir y mantener conexiones, definiendo qu\u00e9 informaci\u00f3n se puede acceder y determinando qui\u00e9n puede acceder a ella.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">El trabajo cambia de extraer informes repetidamente a gestionar y curar el acceso.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">El tercer impacto es estrat\u00e9gico: menos dependencia del proveedor.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">Una integraci\u00f3n construida alrededor de un est\u00e1ndar abierto no necesita necesariamente ser reconstruida si la empresa cambia de herramientas de inteligencia artificial, siempre y cuando la nueva herramienta tambi\u00e9n admita MCP. En un mercado donde los modelos de inteligencia artificial pueden cambiar cada pocos meses, esa flexibilidad puede ayudar a proteger las inversiones tecnol\u00f3gicas.<\/p>\n\n\n\n<div style=\"height:25px\" aria-hidden=\"true\" class=\"wp-block-spacer\"><\/div>\n\n\n\n<h2 class=\"wp-block-heading\">Qu\u00e9 no hace MCP<\/h2>\n\n\n\n<p class=\"wp-block-paragraph\">Sin embargo, hay una expectativa que necesita ser corregida.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">MCP no descubre datos por s\u00ed solo, ni crea integraciones autom\u00e1ticamente. Alguien con experiencia t\u00e9cnica todav\u00eda necesita construir cada conexi\u00f3n y decidir qu\u00e9 puede acceder la IA y qu\u00e9 acciones est\u00e1 permitida a realizar.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">Pero el l\u00edmite real no est\u00e1 definido solo por las funciones disponibles. Tambi\u00e9n depende de las credenciales que el servidor utiliza para acceder a cada sistema. Un servidor con permisos amplios que sirve a un usuario que normalmente no tendr\u00eda esos permisos representa un riesgo de seguridad cl\u00e1sico.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">Eso hace que MCP sea un tema de gobernanza, no solo un tema de tecnolog\u00eda.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">Cuatro \u00e1reas merecen una atenci\u00f3n particular por parte del liderazgo:<\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li><strong>Accuracy.<\/strong> AI interprets data and can still make mistakes or generate inaccurate information. Figures used to support important business decisions should be carefully verified.<\/li>\n\n\n\n<li><strong>Privacy and compliance.<\/strong> Depending on how an AI tool is purchased and configured, strategic company information may pass through external services. These data flows must comply with internal policies and applicable regulations, including General Data Protection Regulation, known as GDPR, particularly in highly regulated industries such as financial services.<\/li>\n\n\n\n<li><strong>Untrusted content.<\/strong> When AI consumes external data, such as social media posts, it may encounter malicious instructions designed to manipulate its behavior. This technique, known as <em>indirect prompt injection<\/em>, makes it especially important to configure integrations according to the principle of least privilege.<\/li>\n\n\n\n<li><strong>Server provenance.<\/strong> An MCP server operates with a high level of trust within an AI workflow, so installing one is similar to granting access to an internal system. In September 2025, a malicious package impersonating the official library of the email service Postmark reportedly began secretly copying messages sent by AI assistants that used it. The lesson is straightforward: the same thousands of public servers that demonstrate adoption also create a broader attack surface. Every MCP server should be evaluated with the same scrutiny applied to any other software vendor.<\/li>\n<\/ul>\n\n\n\n<div style=\"height:25px\" aria-hidden=\"true\" class=\"wp-block-spacer\"><\/div>\n\n\n\n<h2 class=\"wp-block-heading\">La pregunta correcta<\/h2>\n\n\n\n<p class=\"wp-block-paragraph\">Durante a\u00f1os, la promesa de \u201cdemocratizar los datos\u201d enfrent\u00f3 una barrera pr\u00e1ctica: para comunicarse con los datos, las personas primero ten\u00edan que entender el lenguaje de los sistemas que los almacenaban.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">MCP elimina gran parte de esa barrera, pero no resuelve la otra mitad del problema. Si t\u00e9rminos como \u201ccliente activo\u201d o \u201creclamo\u201d significan cosas diferentes en diferentes sistemas, la IA puede proporcionar respuestas r\u00e1pidas y confiadas mientras que dos personas siguen recibiendo respuestas diferentes a la misma pregunta. <\/p>\n\n\n\n<p class=\"wp-block-paragraph\">Establecer la conexi\u00f3n es la parte m\u00e1s f\u00e1cil. Establecer el significado sigue siendo una responsabilidad humana.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">Para las empresas, la pregunta ya no es si la IA acceder\u00e1 a sus sistemas. Las preguntas m\u00e1s importantes son qu\u00e9 datos deber\u00eda poder alcanzar, qu\u00e9 reglas deber\u00edan gobernar ese acceso y qui\u00e9n deber\u00eda ser responsable de ello.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">Las empresas que responden bien a esas preguntas pueden acortar significativamente la distancia entre la incertidumbre y la toma de decisiones.<\/p>\n\n\n\n<hr class=\"wp-block-separator has-alpha-channel-opacity\">\n\n\n\n<p class=\"wp-block-paragraph\"><strong>Lea m\u00e1s sobre la IA: <\/strong><\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><a href=\"https:\/\/loxias.ai\/es\/blog\/how-ai-agents-and-protocols-like-mcp-acp-a2a-and-ap2are-building-the-agentic-web\/\" data-type=\"link\" data-id=\"https:\/\/loxias.ai\/blog\/how-ai-agents-and-protocols-like-mcp-acp-a2a-and-ap2are-building-the-agentic-web\/\">C\u00f3mo los agentes de IA y protocolos como MCP, ACP, A2A y AP2Are est\u00e1n construyendo la web agente<\/a><\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><a href=\"https:\/\/loxias.ai\/es\/blog\/acp-and-a2a-communication-language-for-ai-agents\/\" data-type=\"link\" data-id=\"https:\/\/loxias.ai\/blog\/acp-and-a2a-communication-language-for-ai-agents\/\">Protocolo de Comunicaci\u00f3n de Agentes (ACP) y Agente a Agente (A2A): El nuevo lenguaje de comunicaci\u00f3n para agentes de IA<\/a><\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><\/p>\n","protected":false},"excerpt":{"rendered":"<p>Learn how Model Context Protocol, or MCP, connects generative AI to enterprise data, improves access to business insights, and introduces new governance and security considerations. An open standard designed to connect generative AI to enterprise systems is changing who within companies can ask questions of their data and how quickly they can get answers. It [&hellip;]<\/p>\n","protected":false},"author":7,"featured_media":22656,"comment_status":"closed","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"_acf_changed":false,"inline_featured_image":false,"footnotes":""},"categories":[48],"tags":[408,405,406,49,178,409,407,411,403,404,79,108,265,410,62],"class_list":["post-22651","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-artificial-intelligence","tag-ai-governance","tag-ai-integration","tag-ai-security","tag-artificial-intelligence","tag-business-intelligence","tag-data-governance","tag-data-integration","tag-digital-transformation","tag-enterprise-ai","tag-enterprise-data","tag-generative-ai","tag-mcp","tag-model-context-protocol","tag-prompt-injection","tag-social-listening"],"acf":[],"_links":{"self":[{"href":"https:\/\/loxias.ai\/wp-json\/wp\/v2\/posts\/22651","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/loxias.ai\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/loxias.ai\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/loxias.ai\/wp-json\/wp\/v2\/users\/7"}],"replies":[{"embeddable":true,"href":"https:\/\/loxias.ai\/wp-json\/wp\/v2\/comments?post=22651"}],"version-history":[{"count":6,"href":"https:\/\/loxias.ai\/wp-json\/wp\/v2\/posts\/22651\/revisions"}],"predecessor-version":[{"id":22673,"href":"https:\/\/loxias.ai\/wp-json\/wp\/v2\/posts\/22651\/revisions\/22673"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/loxias.ai\/wp-json\/wp\/v2\/media\/22656"}],"wp:attachment":[{"href":"https:\/\/loxias.ai\/wp-json\/wp\/v2\/media?parent=22651"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/loxias.ai\/wp-json\/wp\/v2\/categories?post=22651"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/loxias.ai\/wp-json\/wp\/v2\/tags?post=22651"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}