{"id":6233,"date":"2026-11-08T13:49:11","date_gmt":"2026-11-08T13:49:11","guid":{"rendered":"https:\/\/www.arivonix.ai\/blog\/?p=6233"},"modified":"2026-11-09T04:06:59","modified_gmt":"2026-11-09T04:06:59","slug":"data-products-for-ai-agents","status":"publish","type":"post","link":"https:\/\/www.arivonix.ai\/blog\/data-products-for-ai-agents\/","title":{"rendered":"Data Products for AI Agents: Why Analysis-Era Designs Are Being Rebuilt"},"content":{"rendered":"\t\t<div data-elementor-type=\"wp-post\" data-elementor-id=\"6233\" class=\"elementor elementor-6233\">\n\t\t\t\t<div class=\"elementor-element elementor-element-3407205 e-flex e-con-boxed e-con e-parent\" data-id=\"3407205\" data-element_type=\"container\" data-e-type=\"container\">\n\t\t\t\t\t<div class=\"e-con-inner\">\n\t\t\t\t<div class=\"elementor-element elementor-element-2fb1f62 elementor-widget elementor-widget-text-editor\" data-id=\"2fb1f62\" data-element_type=\"widget\" data-e-type=\"widget\" data-widget_type=\"text-editor.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t\t\t\t\t\t\t<p>For most of the data mesh era, a data product had a predictable reader. An analyst opened the catalog entry, skimmed the documentation, noticed that the revenue figure excluded refunds, and messaged the owner when something looked off. The product worked because a person filled in what it left unsaid.\u00a0<\/p><p>That reader now has company. Over the past year, <a href=\"https:\/\/docs.snowflake.com\/en\/user-guide\/snowflake-cortex\/cortex-agents-mcp\" target=\"_blank\" rel=\"noopener\">Snowflake<\/a>, <a href=\"https:\/\/www.databricks.com\/blog\/whats-new-unity-catalog-data-ai-summit-2026\" target=\"_blank\" rel=\"noopener\">Databricks<\/a>, and <a href=\"https:\/\/discuss.google.dev\/t\/agentic-data-products-with-knowledge-catalog-and-mcp-remote-server\/367004\" target=\"_blank\" rel=\"noopener\">Google Cloud<\/a> have each given agents a standard route to governed data through the Model Context Protocol (MCP). Data products for AI agents now get called directly and at volume, with nobody in the loop to ask what a column means.\u00a0<\/p><p>The adoption numbers back this up. Fivetran&#8217;s <a href=\"https:\/\/www.fivetran.com\/press\/fivetran-launches-2026-agentic-ai-readiness-index-revealing-gap-between-enterprise-investment-and-data-preparedness-for-agentic-ai\" target=\"_blank\" rel=\"noopener\">2026 Agentic AI Readiness Index<\/a>, a survey of 400 data professionals at mid-sized and large enterprises, found that 41% of organizations already run agentic AI in production. Only 15% say they&#8217;re fully prepared to support it, and data quality and lineage topped the list of barriers at 42%.\u00a0<\/p><p>Analysts have moved too. In May 2026, <a href=\"https:\/\/www.gartner.com\/en\/newsroom\/press-releases\/2026-05-11-gartner-says-lack-of-semantics-causes-inaccurate-artificial-intelligence-agents-and-wasted-spending\" target=\"_blank\" rel=\"noopener\">Gartner predicted<\/a> that by 2027, organizations that prioritize semantics in AI-ready data will improve agentic AI accuracy by up to 80% and cut costs by up to 60%. In July, it followed with a <a href=\"https:\/\/www.gartner.com\/en\/documents\/8150329\" target=\"_blank\" rel=\"noopener\">playbook<\/a> on launching and scaling agentic AI-ready data products.\u00a0<\/p><p>At Arivonix, we see the work ahead as moving promises to a new place. Everything a person used to learn from the docs or a quick message to the owner now has to live inside the product, in a form an agent can read and a platform can check at runtime. That includes definitions, guarantees, permitted uses, and version history.\u00a0<\/p><p>In practice, an agent can only trust what the product declares. The rest of this piece covers what that looks like in 2026 and how we&#8217;d approach the retrofit.\u00a0<\/p><h2><strong>What Data Products for AI Agents Look Like in 2026\u00a0<\/strong><\/h2><p>Data products for AI agents are governed data packages that state their meaning, quality guarantees, permitted uses, and version history in machine-readable form. Agents reach them through a defined interface such as MCP or an API.<\/p>\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-65b39a9 elementor-widget elementor-widget-image\" data-id=\"65b39a9\" data-element_type=\"widget\" data-e-type=\"widget\" data-widget_type=\"image.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t<img fetchpriority=\"high\" decoding=\"async\" width=\"800\" height=\"450\" src=\"https:\/\/www.arivonix.ai\/blog\/wp-content\/uploads\/2026\/11\/01_agent_ready_data_product_layers-1024x576.jpg\" class=\"attachment-large size-large wp-image-6236\" alt=\"agent_ready_data_product_layers\" srcset=\"https:\/\/www.arivonix.ai\/blog\/wp-content\/uploads\/2026\/11\/01_agent_ready_data_product_layers-1024x576.jpg 1024w, https:\/\/www.arivonix.ai\/blog\/wp-content\/uploads\/2026\/11\/01_agent_ready_data_product_layers-300x169.jpg 300w, https:\/\/www.arivonix.ai\/blog\/wp-content\/uploads\/2026\/11\/01_agent_ready_data_product_layers-768x432.jpg 768w, https:\/\/www.arivonix.ai\/blog\/wp-content\/uploads\/2026\/11\/01_agent_ready_data_product_layers-1536x864.jpg 1536w, https:\/\/www.arivonix.ai\/blog\/wp-content\/uploads\/2026\/11\/01_agent_ready_data_product_layers-2048x1152.jpg 2048w\" sizes=\"(max-width: 800px) 100vw, 800px\" title=\"\">\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-185943e elementor-widget elementor-widget-text-editor\" data-id=\"185943e\" data-element_type=\"widget\" data-e-type=\"widget\" data-widget_type=\"text-editor.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t\t\t\t\t\t\t<p><span data-contrast=\"auto\">Google Cloud <\/span><a href=\"https:\/\/cloud.google.com\/transform\/from-asset-to-action-how-data-products-have-become-the-foundation-for-ai-agents\" target=\"_blank\" rel=\"noopener\"><span data-contrast=\"none\">describes the data product<\/span><\/a><span data-contrast=\"auto\"> as the container that packages data, semantics, and governance together.<\/span><span data-ccp-props=\"{&quot;335559739&quot;:160}\">\u00a0<\/span><\/p><p><span data-contrast=\"auto\">This year&#8217;s releases show what goes in that container. Google&#8217;s <\/span><a href=\"https:\/\/discuss.google.dev\/t\/agentic-data-products-with-knowledge-catalog-and-mcp-remote-server\/367004\" target=\"_blank\" rel=\"noopener\"><span data-contrast=\"none\">Knowledge Catalog data products<\/span><\/a><span data-contrast=\"auto\"> bundle tables and views with documentation, sample queries, and access groups, plus contracts that can state refresh cadence and quality standards. Databricks introduced <\/span><a href=\"https:\/\/www.databricks.com\/blog\/whats-new-unity-catalog-data-ai-summit-2026\" target=\"_blank\" rel=\"noopener\"><span data-contrast=\"none\">Domains<\/span><\/a><span data-contrast=\"auto\">, now in public preview, which people browse and agents query through an internal marketplace.<\/span><span data-ccp-props=\"{&quot;335559739&quot;:160}\">\u00a0<\/span><\/p><p><span data-contrast=\"auto\">The principles of data as a product still hold. Clear ownership and service levels matter as much as they did before. The difference is that each one now has to be written down so software can act on it without a follow-up question.<\/span><span data-ccp-props=\"{&quot;335559739&quot;:160}\">\u00a0<\/span><\/p><h2 aria-level=\"2\"><b><span data-contrast=\"none\">How AI Agents Consume Data Products Differently From Analysts<\/span><\/b><span data-ccp-props=\"{&quot;134245418&quot;:true,&quot;335559738&quot;:320,&quot;335559739&quot;:140}\">\u00a0<\/span><\/h2><p><span data-contrast=\"auto\">Analysis-era data products assumed a human would supply missing context. Agents can&#8217;t do that. They pick tools at runtime, chain one result into the next step, and run the same query thousands of times.<\/span><span data-ccp-props=\"{&quot;335559739&quot;:160}\">\u00a0<\/span><\/p><p><span data-contrast=\"auto\">So a gap that one analyst would have caught once gets repeated in every run.<\/span><\/p>\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-ac4d697 elementor-widget elementor-widget-image\" data-id=\"ac4d697\" data-element_type=\"widget\" data-e-type=\"widget\" data-widget_type=\"image.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t<img decoding=\"async\" width=\"800\" height=\"400\" src=\"https:\/\/www.arivonix.ai\/blog\/wp-content\/uploads\/2026\/11\/02_analyst_vs_agent_missing_definition-1024x512.jpg\" class=\"attachment-large size-large wp-image-6237\" alt=\"analyst_vs_agent_missing_definition\" srcset=\"https:\/\/www.arivonix.ai\/blog\/wp-content\/uploads\/2026\/11\/02_analyst_vs_agent_missing_definition-1024x512.jpg 1024w, https:\/\/www.arivonix.ai\/blog\/wp-content\/uploads\/2026\/11\/02_analyst_vs_agent_missing_definition-300x150.jpg 300w, https:\/\/www.arivonix.ai\/blog\/wp-content\/uploads\/2026\/11\/02_analyst_vs_agent_missing_definition-768x384.jpg 768w, https:\/\/www.arivonix.ai\/blog\/wp-content\/uploads\/2026\/11\/02_analyst_vs_agent_missing_definition-1536x768.jpg 1536w, https:\/\/www.arivonix.ai\/blog\/wp-content\/uploads\/2026\/11\/02_analyst_vs_agent_missing_definition-2048x1024.jpg 2048w\" sizes=\"(max-width: 800px) 100vw, 800px\" title=\"\">\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-22c8b43 elementor-widget elementor-widget-text-editor\" data-id=\"22c8b43\" data-element_type=\"widget\" data-e-type=\"widget\" data-widget_type=\"text-editor.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t\t\t\t\t\t\t<p>Builders already feel this. In Monte Carlo&#8217;s <a href=\"https:\/\/info.montecarlodata.com\/monte-carlo-agents-in-production-the-builders-perspective\" target=\"_blank\" rel=\"noopener\">2026 Agents in Production survey<\/a> of 260 AI practitioners, 64% named context and data quality as their biggest visibility blind spot. Only 47% said their systems were easy to trace end to end when something went wrong.<\/p>\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-418a61a elementor-widget elementor-widget-html\" data-id=\"418a61a\" data-element_type=\"widget\" data-e-type=\"widget\" data-widget_type=\"html.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t\t\t\r\n<!-- MODERN ANALYST-ERA VS AGENT-READY DATA PRODUCT TABLE -->\r\n\r\n<div class=\"modern-data-products\">\r\n  <div class=\"mdp-top-accent\"><\/div>\r\n\r\n  <div class=\"mdp-table-container\">\r\n    <table class=\"mdp-table\" aria-label=\"Analyst-era versus agent-ready data products\">\r\n\r\n      <thead>\r\n        <tr>\r\n          <th>Dimension<\/th>\r\n          <th>Analyst-era data product<\/th>\r\n          <th>Agent-ready data product<\/th>\r\n        <\/tr>\r\n      <\/thead>\r\n\r\n      <tbody>\r\n        <tr>\r\n          <td data-label=\"Dimension\">\r\n            <span class=\"mdp-label\">Meaning<\/span>\r\n          <\/td>\r\n          <td data-label=\"Analyst-era data product\">\r\n            Documentation, glossary pages, and tribal knowledge\r\n          <\/td>\r\n          <td data-label=\"Agent-ready data product\" class=\"mdp-modern-cell\">\r\n            Definitions and metric logic declared in a semantic layer the agent queries\r\n          <\/td>\r\n        <\/tr>\r\n\r\n        <tr>\r\n          <td data-label=\"Dimension\">\r\n            <span class=\"mdp-label\">Trust signal<\/span>\r\n          <\/td>\r\n          <td data-label=\"Analyst-era data product\">\r\n            Owner reputation and dashboard history\r\n          <\/td>\r\n          <td data-label=\"Agent-ready data product\" class=\"mdp-modern-cell\">\r\n            Freshness, quality, and certification status readable at query time\r\n          <\/td>\r\n        <\/tr>\r\n\r\n        <tr>\r\n          <td data-label=\"Dimension\">\r\n            <span class=\"mdp-label\">Access<\/span>\r\n          <\/td>\r\n          <td data-label=\"Analyst-era data product\">\r\n            Requested once, then used broadly\r\n          <\/td>\r\n          <td data-label=\"Agent-ready data product\" class=\"mdp-modern-cell\">\r\n            Tied to the agent's identity and a declared purpose\r\n          <\/td>\r\n        <\/tr>\r\n\r\n        <tr>\r\n          <td data-label=\"Dimension\">\r\n            <span class=\"mdp-label\">Change<\/span>\r\n          <\/td>\r\n          <td data-label=\"Analyst-era data product\">\r\n            Release notes and a heads-up email\r\n          <\/td>\r\n          <td data-label=\"Agent-ready data product\" class=\"mdp-modern-cell\">\r\n            Versioned definitions with deprecation windows an agent can detect\r\n          <\/td>\r\n        <\/tr>\r\n\r\n        <tr>\r\n          <td data-label=\"Dimension\">\r\n            <span class=\"mdp-label\">Interface<\/span>\r\n          <\/td>\r\n          <td data-label=\"Analyst-era data product\">\r\n            SQL, BI connectors, and file exports\r\n          <\/td>\r\n          <td data-label=\"Agent-ready data product\" class=\"mdp-modern-cell\">\r\n            A small set of MCP tools or APIs with scoped outputs\r\n          <\/td>\r\n        <\/tr>\r\n\r\n        <tr>\r\n          <td data-label=\"Dimension\">\r\n            <span class=\"mdp-label\">Feedback<\/span>\r\n          <\/td>\r\n          <td data-label=\"Analyst-era data product\">\r\n            User complaints and adoption counts\r\n          <\/td>\r\n          <td data-label=\"Agent-ready data product\" class=\"mdp-modern-cell\">\r\n            Per-agent usage telemetry linked to downstream actions\r\n          <\/td>\r\n        <\/tr>\r\n      <\/tbody>\r\n\r\n    <\/table>\r\n  <\/div>\r\n<\/div>\r\n\r\n<style>\r\n\/* =========================================\r\n   PREMIUM DATA PRODUCT COMPARISON\r\n   Fully scoped WordPress CSS\r\n========================================= *\/\r\n\r\n.modern-data-products,\r\n.modern-data-products * {\r\n  box-sizing: border-box !important;\r\n}\r\n\r\n.modern-data-products {\r\n  width: 100% !important;\r\n  max-width: 100% !important;\r\n  margin: 24px 0 !important;\r\n  padding: 0 !important;\r\n  font-family: Arial, Helvetica, sans-serif !important;\r\n  color: #263650 !important;\r\n  font-size: 13px !important;\r\n  line-height: 1.55 !important;\r\n}\r\n\r\n\/* Accent line *\/\r\n\r\n.modern-data-products .mdp-top-accent {\r\n  width: 100% !important;\r\n  height: 4px !important;\r\n  background: linear-gradient(\r\n    90deg,\r\n    #243b80,\r\n    #4264cf,\r\n    #7863ef\r\n  ) !important;\r\n  border-radius: 8px 8px 0 0 !important;\r\n}\r\n\r\n\/* Outer container *\/\r\n\r\n.modern-data-products .mdp-table-container {\r\n  width: 100% !important;\r\n  max-width: 100% !important;\r\n  overflow: hidden !important;\r\n  border: 1px solid #dce5f1 !important;\r\n  border-top: 0 !important;\r\n  border-radius: 0 0 10px 10px !important;\r\n  background: #ffffff !important;\r\n  box-shadow: 0 5px 22px rgba(24, 45, 85, 0.07) !important;\r\n}\r\n\r\n\/* Main table *\/\r\n\r\n.modern-data-products table.mdp-table {\r\n  display: table !important;\r\n  width: 100% !important;\r\n  max-width: 100% !important;\r\n  margin: 0 !important;\r\n  padding: 0 !important;\r\n  border: 0 !important;\r\n  border-collapse: collapse !important;\r\n  border-spacing: 0 !important;\r\n  table-layout: fixed !important;\r\n  background: #ffffff !important;\r\n  color: #263650 !important;\r\n  font: 13px\/1.55 Arial, Helvetica, sans-serif !important;\r\n  box-shadow: none !important;\r\n}\r\n\r\n\/* Header *\/\r\n\r\n.modern-data-products .mdp-table thead {\r\n  display: table-header-group !important;\r\n}\r\n\r\n.modern-data-products .mdp-table thead tr {\r\n  display: table-row !important;\r\n}\r\n\r\n.modern-data-products .mdp-table thead th {\r\n  display: table-cell !important;\r\n  padding: 16px 14px !important;\r\n  border: 0 !important;\r\n  border-right: 1px solid rgba(255,255,255,.13) !important;\r\n  border-bottom: 2px solid #728cf5 !important;\r\n  background: linear-gradient(\r\n    115deg,\r\n    #172b59 0%,\r\n    #213d7c 55%,\r\n    #393fa9 100%\r\n  ) !important;\r\n  color: #ffffff !important;\r\n  text-align: left !important;\r\n  vertical-align: middle !important;\r\n  font: 700 13px\/1.4 Arial, Helvetica, sans-serif !important;\r\n  letter-spacing: 0 !important;\r\n  text-transform: none !important;\r\n  white-space: normal !important;\r\n  box-shadow: none !important;\r\n}\r\n\r\n\/* Column widths *\/\r\n\r\n.modern-data-products .mdp-table thead th:nth-child(1) {\r\n  width: 19% !important;\r\n}\r\n\r\n.modern-data-products .mdp-table thead th:nth-child(2) {\r\n  width: 40% !important;\r\n}\r\n\r\n.modern-data-products .mdp-table thead th:nth-child(3) {\r\n  width: 41% !important;\r\n}\r\n\r\n\/* Table rows *\/\r\n\r\n.modern-data-products .mdp-table tbody {\r\n  display: table-row-group !important;\r\n}\r\n\r\n.modern-data-products .mdp-table tbody tr {\r\n  display: table-row !important;\r\n}\r\n\r\n\/* Table cells *\/\r\n\r\n.modern-data-products .mdp-table tbody td {\r\n  display: table-cell !important;\r\n  padding: 14px 14px !important;\r\n  border: 0 !important;\r\n  border-right: 1px solid #e7ecf4 !important;\r\n  border-bottom: 1px solid #e7ecf4 !important;\r\n  background: #ffffff !important;\r\n  color: #45536a !important;\r\n  text-align: left !important;\r\n  vertical-align: middle !important;\r\n  font: 400 13px\/1.55 Arial, Helvetica, sans-serif !important;\r\n  overflow-wrap: anywhere !important;\r\n  word-break: normal !important;\r\n  white-space: normal !important;\r\n  box-shadow: none !important;\r\n}\r\n\r\n\/* Alternating backgrounds *\/\r\n\r\n.modern-data-products .mdp-table tbody tr:nth-child(even) td {\r\n  background: #f5f8fd !important;\r\n}\r\n\r\n\/* Dimension labels *\/\r\n\r\n.modern-data-products .mdp-table tbody td:first-child {\r\n  background: #edf3fc !important;\r\n  color: #172b59 !important;\r\n  font-weight: 700 !important;\r\n}\r\n\r\n.modern-data-products .mdp-label {\r\n  display: inline-block !important;\r\n  color: #1d376e !important;\r\n  font-size: 12px !important;\r\n  font-weight: 700 !important;\r\n  line-height: 1.5 !important;\r\n}\r\n\r\n\/* Highlight agent-ready information *\/\r\n\r\n.modern-data-products .mdp-table tbody td.mdp-modern-cell {\r\n  color: #263e79 !important;\r\n  font-weight: 500 !important;\r\n}\r\n\r\n\/* Remove final borders *\/\r\n\r\n.modern-data-products .mdp-table tbody td:last-child {\r\n  border-right: 0 !important;\r\n}\r\n\r\n.modern-data-products .mdp-table tbody tr:last-child td {\r\n  border-bottom: 0 !important;\r\n}\r\n\r\n\/* Hover *\/\r\n\r\n@media (hover: hover) {\r\n  .modern-data-products .mdp-table tbody tr:hover td {\r\n    background: #eaf1ff !important;\r\n    transition: background 0.2s ease !important;\r\n  }\r\n}\r\n\r\n\/* =========================================\r\n   TABLET\r\n========================================= *\/\r\n\r\n@media screen and (min-width: 601px) and (max-width: 900px) {\r\n\r\n  .modern-data-products .mdp-table thead th {\r\n    padding: 12px 10px !important;\r\n    font-size: 12px !important;\r\n  }\r\n\r\n  .modern-data-products .mdp-table tbody td {\r\n    padding: 11px 10px !important;\r\n    font-size: 12px !important;\r\n  }\r\n\r\n  .modern-data-products .mdp-label {\r\n    font-size: 11px !important;\r\n  }\r\n}\r\n\r\n\/* =========================================\r\n   MOBILE - CARD LAYOUT\r\n========================================= *\/\r\n\r\n@media screen and (max-width: 600px) {\r\n\r\n  .modern-data-products {\r\n    margin: 16px 0 !important;\r\n  }\r\n\r\n  .modern-data-products .mdp-table-container {\r\n    overflow: visible !important;\r\n    border: 0 !important;\r\n    border-radius: 0 !important;\r\n    box-shadow: none !important;\r\n    background: transparent !important;\r\n  }\r\n\r\n  .modern-data-products .mdp-table,\r\n  .modern-data-products .mdp-table tbody,\r\n  .modern-data-products .mdp-table tbody tr,\r\n  .modern-data-products .mdp-table tbody td {\r\n    display: block !important;\r\n    width: 100% !important;\r\n    max-width: 100% !important;\r\n  }\r\n\r\n  .modern-data-products .mdp-table {\r\n    table-layout: auto !important;\r\n    border: 0 !important;\r\n    background: transparent !important;\r\n  }\r\n\r\n  \/* Visually hide desktop headings *\/\r\n\r\n  .modern-data-products .mdp-table thead {\r\n    display: block !important;\r\n    position: absolute !important;\r\n    width: 1px !important;\r\n    height: 1px !important;\r\n    margin: -1px !important;\r\n    padding: 0 !important;\r\n    overflow: hidden !important;\r\n    clip: rect(0,0,0,0) !important;\r\n    white-space: nowrap !important;\r\n    border: 0 !important;\r\n  }\r\n\r\n  .modern-data-products .mdp-table tbody {\r\n    display: block !important;\r\n  }\r\n\r\n  \/* Card design *\/\r\n\r\n  .modern-data-products .mdp-table tbody tr {\r\n    display: block !important;\r\n    margin: 0 0 15px !important;\r\n    padding: 0 !important;\r\n    border: 1px solid #dce5f1 !important;\r\n    border-radius: 10px !important;\r\n    overflow: hidden !important;\r\n    background: #ffffff !important;\r\n    box-shadow: 0 4px 14px rgba(28,53,97,.06) !important;\r\n  }\r\n\r\n  .modern-data-products .mdp-table tbody td {\r\n    display: block !important;\r\n    width: 100% !important;\r\n    padding: 12px 15px !important;\r\n    border: 0 !important;\r\n    border-bottom: 1px solid #e8edf5 !important;\r\n    background: #ffffff !important;\r\n    color: #45536a !important;\r\n    font-size: 13px !important;\r\n    line-height: 1.55 !important;\r\n    text-align: left !important;\r\n  }\r\n\r\n  \/* Dimension card heading *\/\r\n\r\n  .modern-data-products .mdp-table tbody td:first-child {\r\n    background: linear-gradient(\r\n      100deg,\r\n      #edf3fc,\r\n      #e7edff\r\n    ) !important;\r\n    border-top: 3px solid #4355ca !important;\r\n    color: #172b59 !important;\r\n  }\r\n\r\n  \/* Field labels *\/\r\n\r\n  .modern-data-products .mdp-table tbody td::before {\r\n    content: attr(data-label) !important;\r\n    display: block !important;\r\n    margin: 0 0 5px !important;\r\n    color: #71809a !important;\r\n    font: 700 10px\/1.4 Arial, Helvetica, sans-serif !important;\r\n    letter-spacing: .07em !important;\r\n    text-transform: uppercase !important;\r\n  }\r\n\r\n  .modern-data-products .mdp-table tbody td:last-child {\r\n    border-bottom: 0 !important;\r\n  }\r\n\r\n  \/* Reset desktop alternating rows *\/\r\n\r\n  .modern-data-products .mdp-table tbody tr:nth-child(even) td {\r\n    background: #ffffff !important;\r\n  }\r\n\r\n  .modern-data-products .mdp-table tbody tr:nth-child(even) td:first-child {\r\n    background: linear-gradient(\r\n      100deg,\r\n      #edf3fc,\r\n      #e7edff\r\n    ) !important;\r\n  }\r\n\r\n  .modern-data-products .mdp-table tbody td.mdp-modern-cell {\r\n    color: #263e79 !important;\r\n  }\r\n}\r\n<\/style>\r\n\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-0e7c19c elementor-widget elementor-widget-text-editor\" data-id=\"0e7c19c\" data-element_type=\"widget\" data-e-type=\"widget\" data-widget_type=\"text-editor.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t\t\t\t\t\t\t<p>Source: Arivonix analysis.\u00a0<\/p><p><span data-contrast=\"auto\">The access row is usually where projects stall. Most entitlement models were written for people and service accounts, and an agent acting on someone&#8217;s behalf fits neither cleanly.<\/span><span data-ccp-props=\"{&quot;335559739&quot;:160}\">\u00a0<\/span><\/p><p><span data-contrast=\"auto\">Our analysis of <\/span><a href=\"https:\/\/www.arivonix.ai\/blog\/ai-agent-identity\/\" target=\"_blank\" rel=\"noopener\"><span data-contrast=\"none\">AI agent identity<\/span><\/a><span data-contrast=\"auto\"> covers why credentials alone don&#8217;t settle what data an agent should see.<\/span><span data-ccp-props=\"{&quot;335559739&quot;:160}\">\u00a0<\/span><\/p><h2 aria-level=\"2\"><b><span data-contrast=\"none\">AI-Ready Data Products Are Moving Into Code<\/span><\/b><span data-ccp-props=\"{&quot;134245418&quot;:true,&quot;335559738&quot;:320,&quot;335559739&quot;:140}\">\u00a0<\/span><\/h2><p><span data-contrast=\"auto\">The biggest shift this year is that the parts of a data product people used to read are turning into versioned specifications that machines can parse.<\/span><span data-ccp-props=\"{&quot;335559739&quot;:160}\">\u00a0<\/span><\/p><ul><li aria-setsize=\"-1\" data-leveltext=\"\u2022\" data-font=\"\" data-listid=\"2\" data-list-defn-props=\"{&quot;335552541&quot;:1,&quot;335559685&quot;:720,&quot;335559991&quot;:360,&quot;469769242&quot;:[8226],&quot;469777803&quot;:&quot;left&quot;,&quot;469777804&quot;:&quot;\u2022&quot;,&quot;469777815&quot;:&quot;hybridMultilevel&quot;}\" data-aria-posinset=\"1\" data-aria-level=\"1\"><b><span data-contrast=\"auto\">Semantics.<\/span><\/b><span data-contrast=\"auto\"> The <\/span><a href=\"https:\/\/www.snowflake.com\/en\/blog\/open-semantic-interchanges-specs-finalized\/\" target=\"_blank\" rel=\"noopener\"><span data-contrast=\"none\">Open Semantic Interchange (OSI) specification<\/span><\/a><span data-contrast=\"auto\"> published its first version in January 2026 as a vendor-neutral format for metrics, dimensions, and relationships, and Databricks joined the working group. In July, the project entered the Apache Incubator under a new name, <\/span><a href=\"https:\/\/www.snowflake.com\/en\/blog\/apache-ossie-open-semantic-interchange-incubator\/\" target=\"_blank\" rel=\"noopener\"><span data-contrast=\"none\">Apache Ossie<\/span><\/a><span data-contrast=\"auto\">. Databricks&#8217; Unity Catalog Metrics, launched in June, are defined once, can be queried from SQL, BI tools, APIs, and agents, and are labeled OSI-ready. Portability is still early, though. Reference converters for dbt and Apache Polaris exist, but native import and export across the major platforms remains thin.<\/span><span data-ccp-props=\"{&quot;335559739&quot;:100}\">\u00a0<\/span><\/li><\/ul><ul><li aria-setsize=\"-1\" data-leveltext=\"\u2022\" data-font=\"\" data-listid=\"2\" data-list-defn-props=\"{&quot;335552541&quot;:1,&quot;335559685&quot;:720,&quot;335559991&quot;:360,&quot;469769242&quot;:[8226],&quot;469777803&quot;:&quot;left&quot;,&quot;469777804&quot;:&quot;\u2022&quot;,&quot;469777815&quot;:&quot;hybridMultilevel&quot;}\" data-aria-posinset=\"2\" data-aria-level=\"1\"><b><span data-contrast=\"auto\">Contracts.<\/span><\/b><span data-contrast=\"auto\"> The Linux Foundation&#8217;s Bitol project maintains both the <\/span><a href=\"https:\/\/github.com\/bitol-io\/open-data-contract-standard\" target=\"_blank\" rel=\"noopener\"><span data-contrast=\"none\">Open Data Contract Standard<\/span><\/a><span data-contrast=\"auto\"> and the <\/span><a href=\"https:\/\/bitol.io\/announcing-odps-v1-0-0-building-the-language-of-data-products\/\" target=\"_blank\" rel=\"noopener\"><span data-contrast=\"none\">Open Data Product Standard<\/span><\/a><span data-contrast=\"auto\">, which reached version 1.0 in September 2025. Together they give teams one machine-readable way to declare a product and the guarantees attached to it. Gartner&#8217;s playbook lists machine-verifiable data contracts among the practices that make data products ready for agentic AI.<\/span><span data-ccp-props=\"{&quot;335559739&quot;:100}\">\u00a0<\/span><\/li><\/ul><ul><li aria-setsize=\"-1\" data-leveltext=\"\u2022\" data-font=\"\" data-listid=\"2\" data-list-defn-props=\"{&quot;335552541&quot;:1,&quot;335559685&quot;:720,&quot;335559991&quot;:360,&quot;469769242&quot;:[8226],&quot;469777803&quot;:&quot;left&quot;,&quot;469777804&quot;:&quot;\u2022&quot;,&quot;469777815&quot;:&quot;hybridMultilevel&quot;}\" data-aria-posinset=\"3\" data-aria-level=\"1\"><b><span data-contrast=\"auto\">Metadata as a release artifact.<\/span><\/b><span data-contrast=\"auto\"> When <\/span><a href=\"https:\/\/www.starburst.io\/blog\/what-do-agents-need-from-data-products\/\" target=\"_blank\" rel=\"noopener\"><span data-contrast=\"none\">Starburst ran its own analytics agent<\/span><\/a><span data-contrast=\"auto\"> for company leadership, one lesson was to version metadata like source code, so every answer can be traced to the definition it used. Deprecated versions get retired and hidden from agents.<\/span><span data-ccp-props=\"{&quot;335559739&quot;:100}\">\u00a0<\/span><\/li><\/ul><p><span data-contrast=\"auto\">The teams furthest along handle a change to a business definition the way they&#8217;d handle a code change. It gets reviewed and versioned before it&#8217;s promoted.<\/span><span data-ccp-props=\"{&quot;335559739&quot;:160}\">\u00a0<\/span><\/p><p><span data-contrast=\"auto\">The open problem is semantic sprawl. An agent working across two platforms can still meet two definitions of the same metric, and no interchange format settles that yet.<\/span><span data-ccp-props=\"{&quot;335559739&quot;:160}\">\u00a0<\/span><\/p><p><span data-contrast=\"auto\">For relationship-heavy domains, some teams add a knowledge graph next to the semantic layer, an approach we compared in <\/span><a href=\"https:\/\/www.arivonix.ai\/blog\/graphrag-vs-standard-rag-enterprise-ai\/\" target=\"_blank\" rel=\"noopener\"><span data-contrast=\"none\">GraphRAG vs standard RAG<\/span><\/a><span data-contrast=\"auto\">.<\/span><span data-ccp-props=\"{&quot;335559739&quot;:160}\">\u00a0<\/span><\/p><h2 aria-level=\"2\"><b><span data-contrast=\"none\">MCP Turned Data Products Into Agent Interfaces<\/span><\/b><span data-ccp-props=\"{&quot;134245418&quot;:true,&quot;335559738&quot;:320,&quot;335559739&quot;:140}\">\u00a0<\/span><\/h2><p><span data-contrast=\"auto\">A year ago, connecting an agent to a data product usually meant a custom integration. In 2026, MCP became the common way in.<\/span><span data-ccp-props=\"{&quot;335559739&quot;:160}\">\u00a0<\/span><\/p><p><span data-contrast=\"auto\">Snowflake&#8217;s managed MCP server exposes semantic views and search services as tools. Its <\/span><a href=\"https:\/\/www.snowflake.com\/en\/developers\/guides\/getting-started-with-snowflake-mcp-server\/\" target=\"_blank\" rel=\"noopener\"><span data-contrast=\"none\">setup guide<\/span><\/a><span data-contrast=\"auto\"> makes a point worth copying: access to the server doesn&#8217;t grant access to its tools, because the connecting role needs privileges on each underlying object. Snowflake extended the pattern to <\/span><a href=\"https:\/\/docs.snowflake.com\/release-notes\/2026\/other\/2026-08-07-native-apps-agents-mcp-ga\" target=\"_blank\" rel=\"noopener\"><span data-contrast=\"none\">Native Apps<\/span><\/a><span data-contrast=\"auto\"> when agent and MCP support went GA in August 2026.<\/span><span data-ccp-props=\"{&quot;335559739&quot;:160}\">\u00a0<\/span><\/p><p><span data-contrast=\"auto\">Since late May 2026, Google has let agents <\/span><a href=\"https:\/\/discuss.google.dev\/t\/agentic-data-products-with-knowledge-catalog-and-mcp-remote-server\/367004\" target=\"_blank\" rel=\"noopener\"><span data-contrast=\"none\">list, create, and update<\/span><\/a><span data-contrast=\"auto\"> Knowledge Catalog data products through MCP tools. Agents can now produce data products as well as consume them. Databricks exposes Unity Catalog business semantics over MCP and captures model and MCP activity in one tracing layer.<\/span><span data-ccp-props=\"{&quot;335559739&quot;:160}\">\u00a0<\/span><\/p><p><span data-contrast=\"auto\">A research paper on <\/span><a href=\"https:\/\/arxiv.org\/abs\/2601.08687\" target=\"_blank\" rel=\"noopener\"><span data-contrast=\"none\">Data Product MCP<\/span><\/a><span data-contrast=\"auto\">, first published in January 2026, shows where this could go. Agents find, request, and query data products while contracts are enforced in real time, and each generated query is checked against approved business purposes before it runs.<\/span><span data-ccp-props=\"{&quot;335559739&quot;:160}\">\u00a0<\/span><\/p><p><span data-contrast=\"auto\">For MCP data products, the tool list is the product&#8217;s front door. A product owner now decides which questions an agent can ask, on top of which tables it can read. Hand an agent a generic SQL tool and the ambiguity the product was built to remove comes right back.<\/span><span data-ccp-props=\"{&quot;335559739&quot;:160}\">\u00a0<\/span><\/p><p><span data-contrast=\"auto\">Data Minded, which built an MCP server for its own <\/span><a href=\"https:\/\/www.dataminded.com\/resources\/from-idea-to-implementation-building-an-mcp-server-for-the-data-product-portal\" target=\"_blank\" rel=\"noopener\"><span data-contrast=\"none\">data product portal<\/span><\/a><span data-contrast=\"auto\">, advises picking the three to five most valuable tools and getting those working before exposing anything else.<\/span><\/p>\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-52e2498 elementor-widget elementor-widget-image\" data-id=\"52e2498\" data-element_type=\"widget\" data-e-type=\"widget\" data-widget_type=\"image.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t<img decoding=\"async\" width=\"800\" height=\"420\" src=\"https:\/\/www.arivonix.ai\/blog\/wp-content\/uploads\/2026\/11\/04_mcp_tool_surface_architecture-1024x538.jpg\" class=\"attachment-large size-large wp-image-6238\" alt=\"\" srcset=\"https:\/\/www.arivonix.ai\/blog\/wp-content\/uploads\/2026\/11\/04_mcp_tool_surface_architecture-1024x538.jpg 1024w, https:\/\/www.arivonix.ai\/blog\/wp-content\/uploads\/2026\/11\/04_mcp_tool_surface_architecture-300x158.jpg 300w, https:\/\/www.arivonix.ai\/blog\/wp-content\/uploads\/2026\/11\/04_mcp_tool_surface_architecture-768x403.jpg 768w, https:\/\/www.arivonix.ai\/blog\/wp-content\/uploads\/2026\/11\/04_mcp_tool_surface_architecture-1536x806.jpg 1536w, https:\/\/www.arivonix.ai\/blog\/wp-content\/uploads\/2026\/11\/04_mcp_tool_surface_architecture-2048x1075.jpg 2048w\" sizes=\"(max-width: 800px) 100vw, 800px\" title=\"\">\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-46e2d5f elementor-widget elementor-widget-text-editor\" data-id=\"46e2d5f\" data-element_type=\"widget\" data-e-type=\"widget\" data-widget_type=\"text-editor.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t\t\t\t\t\t\t<h2><strong>How to Design Data Products for AI Agents: Our Approach\u00a0<\/strong><\/h2><p>Retrofitting every product at once spreads effort across products no agent will touch this year. Here&#8217;s how we&#8217;d sequence it.\u00a0<\/p><ul><li><strong>Start from one agent decision.<\/strong> Pick a decision an agent will make, such as routing a submission or flagging an account, and work back to the data products it depends on.\u00a0<\/li><\/ul><ul><li><strong>Run an unasked-question audit.<\/strong> Pull the last few months of questions analysts sent the product owner. Each question that keeps coming up is a field the product needs to declare, whether that&#8217;s what a metric excludes, how fresh the data is, or what it shouldn&#8217;t be used for.\u00a0<\/li><\/ul><ul><li><strong>Declare definitions once and serve them.<\/strong> Put metric and entity definitions in the semantic layer the agent queries. Definitions restated in prompts drift from the source within weeks.\u00a0<\/li><\/ul><ul><li><strong>Keep the tool surface narrow.<\/strong> A handful of purpose-built tools per product, like a governed metric query and a freshness check, leave agents less room to misread the data than open SQL access does.\u00a0<\/li><\/ul><ul><li><strong>Make guarantees checkable.<\/strong> Express freshness, certification, and quality commitments as machine-verifiable data contracts, so an agent can confirm a product is current before it acts. A last-reviewed date on a wiki page can&#8217;t do that job.\u00a0<\/li><\/ul><ul><li><strong>Version with visible deprecation windows.<\/strong> A breaking change to a definition gets a new version. The old one stays readable, with a sunset date the agent can see.\u00a0<\/li><\/ul><ul><li><strong>Instrument consumption by agent.<\/strong> Record which agent called which tool, against which version, and what it did next. In the Monte Carlo survey, <a href=\"https:\/\/www.globenewswire.com\/news-release\/2026\/04\/28\/3282734\/0\/en\/two-thirds-64-of-enterprise-leaders-and-engineers-rushed-ai-agents-before-they-were-ready-and-are-now-paying-the-price.html\" target=\"_blank\" rel=\"noopener\">52% of builders<\/a> said customer complaints are one of the main ways they find out about production issues. Per-agent telemetry can surface a misread definition before a customer does, and it doubles as the product&#8217;s adoption metric.\u00a0<\/li><\/ul><ul><li><strong>Test the product with the agent.<\/strong> Before promoting a new version, run a set of known questions with expected answers against it. We cover where these checks belong in our piece on the <a href=\"https:\/\/www.arivonix.ai\/blog\/agent-harness\/\" target=\"_blank\" rel=\"noopener\">agent harness<\/a>.\u00a0<\/li><\/ul><h2><strong>Data Products for AI Agents in Insurance: One Loss History Product, Two Consumers\u00a0<\/strong><\/h2><p>Here&#8217;s a hypothetical. A commercial lines carrier has published a loss history data product for years. It reports incurred losses by account and policy year, and underwriting analysts use it every day.\u00a0<\/p><p>The analysts know things the product never states. &#8220;Incurred&#8221; here means paid losses plus case reserves, excluding IBNR and allocated loss adjustment expense. The latest policy year is still developing, so they discount it.\u00a0<\/p><p>Now the carrier deploys a triage agent that screens renewals against an appetite rule based on five-year loss ratio.\u00a0<\/p><ul><li><strong>Analysis-era product.<\/strong> The agent reads the incurred column at face value and treats the immature year like the others. It also applies a threshold someone wrote assuming expenses were included. Some accounts get referred that shouldn&#8217;t be, others get cleared that should be referred, and nobody notices until an underwriter questions a referral.\u00a0<\/li><\/ul><ul><li><strong>Agent-ready product.<\/strong> The product declares what incurred includes, the valuation date, a maturity flag for each policy year, and a freshness commitment. Its permitted purpose is renewal triage, and pricing is out of scope. Its tool returns the loss ratio with maturity flags attached, so the agent can send immature-year accounts to a human.\u00a0<\/li><\/ul><p>The quickest readiness test is a back-test. Give the agent 20 renewals an experienced underwriter has already triaged and compare the outcomes.\u00a0<\/p><p>Most disagreements trace back to something the product never declared, and each one becomes another entry in the unasked-question audit. Our notes from <a href=\"https:\/\/www.arivonix.ai\/blog\/itc-vegas-2026-underwriting-ai\/\" target=\"_blank\" rel=\"noopener\">ITC Vegas 2026<\/a> cover where underwriting agents are already holding up in production.\u00a0<\/p><h2><strong>Where Arivonix Fits\u00a0<\/strong><\/h2><p>Most of the work in building data products for AI agents is unglamorous. It means getting definitions and permissions out of people&#8217;s heads and into the product itself.\u00a0<\/p><p>The <a href=\"https:\/\/www.arivonix.ai\/data-products-hub\/\" target=\"_blank\" rel=\"noopener\">Arivonix Data Products Hub<\/a> packages data into governed products and publishes them to a marketplace for internal teams and partners. The <a href=\"https:\/\/www.arivonix.ai\/architecture\/\" target=\"_blank\" rel=\"noopener\">Arivonix platform<\/a> runs inside your own secure environment and connects to 250+ data sources with zero data movement.\u00a0<\/p><p><a href=\"https:\/\/www.arivonix.ai\/agent-arivon\/\" target=\"_blank\" rel=\"noopener\">Agent Arivon<\/a> traces every step for audit, and <a href=\"https:\/\/www.arivonix.ai\/data-centric-ai-assurance\/\" target=\"_blank\" rel=\"noopener\">Data-Centric AI Assurance<\/a> adds end-to-end lineage and guardrails across the products agents consume. For insurers, we apply this to <a href=\"https:\/\/www.arivonix.ai\/insurance_industry\/\" target=\"_blank\" rel=\"noopener\">AI for underwriting and claims<\/a>. Our guide to the <a href=\"https:\/\/www.arivonix.ai\/guide\/ai-agent-platform\/\" target=\"_blank\" rel=\"noopener\">enterprise AI agent platform<\/a> walks through running agents in production from start to finish.\u00a0<\/p><p>Agent traffic against data products will keep growing faster than the teams that maintain them. The products that hold up will be the ones that answer an agent&#8217;s questions before it needs to ask.\u00a0<\/p><p>Want to find out which of your data products agents can use reliably today? Talk to our team, or <a href=\"https:\/\/www.arivonix.ai\/free-trial\/\" target=\"_blank\" rel=\"noopener\">try the platform yourself<\/a>.\u00a0<\/p><p><a href=\"https:\/\/www.arivonix.ai\/book-a-consultation\/\" target=\"_blank\" rel=\"noopener\"><strong>Book a Consultation<\/strong><\/a>\u00a0<\/p>\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t","protected":false},"excerpt":{"rendered":"<p>For most of the data mesh era, a data product had a predictable reader. An analyst opened the catalog entry, skimmed the documentation, noticed that the revenue figure excluded refunds, and messaged the owner when something looked off. The product worked because a person filled in what it left unsaid.\u00a0 That reader now has company. [&hellip;]<\/p>\n","protected":false},"author":9,"featured_media":6235,"comment_status":"closed","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"_acf_changed":false,"footnotes":""},"categories":[140],"tags":[],"class_list":["post-6233","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-arivonix"],"acf":[],"_links":{"self":[{"href":"https:\/\/www.arivonix.ai\/blog\/wp-json\/wp\/v2\/posts\/6233","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.arivonix.ai\/blog\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.arivonix.ai\/blog\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.arivonix.ai\/blog\/wp-json\/wp\/v2\/users\/9"}],"replies":[{"embeddable":true,"href":"https:\/\/www.arivonix.ai\/blog\/wp-json\/wp\/v2\/comments?post=6233"}],"version-history":[{"count":4,"href":"https:\/\/www.arivonix.ai\/blog\/wp-json\/wp\/v2\/posts\/6233\/revisions"}],"predecessor-version":[{"id":6241,"href":"https:\/\/www.arivonix.ai\/blog\/wp-json\/wp\/v2\/posts\/6233\/revisions\/6241"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.arivonix.ai\/blog\/wp-json\/wp\/v2\/media\/6235"}],"wp:attachment":[{"href":"https:\/\/www.arivonix.ai\/blog\/wp-json\/wp\/v2\/media?parent=6233"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.arivonix.ai\/blog\/wp-json\/wp\/v2\/categories?post=6233"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.arivonix.ai\/blog\/wp-json\/wp\/v2\/tags?post=6233"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}