{"templateName":"base-page-template54","cssClassNames":"page basicpage summit-page","allowedRenditionsWidth":["320","480","640","768","960","1200","1440","1920"],"language":"en","description":"Explore AI use cases across 13 industries, from healthcare and finance to retail and manufacturing, and learn what it takes to deploy AI reliably.","title":"AI Use Cases by Industry: Real-World Examples | Snowflake","analyticsPageType":"homepage","analyticsCategory":"general","analyticsSubCategory":"","excludeFromAnalytics":false,":mappedPath":"/en/artificial-intelligence/industries/",":type":"snowflake-site/components/structure/page",":items":{"root":{"columnCount":12,"columnClassNames":{"markup_editor_928258845":"aem-GridColumn aem-GridColumn--default--12","experiencefragment-banner":"aem-GridColumn aem-GridColumn--default--12","experiencefragment-header":"aem-GridColumn aem-GridColumn--default--12","responsivegrid":"aem-GridColumn aem-GridColumn--default--12","markup_editor_597730182":"aem-GridColumn aem-GridColumn--default--12","experiencefragment-footer":"aem-GridColumn aem-GridColumn--default--12","experiencefragment":"aem-GridColumn aem-GridColumn--default--12","modal_container":"aem-GridColumn aem-GridColumn--default--12","markup_editor":"aem-GridColumn aem-GridColumn--default--12"},"gridClassNames":"aem-Grid aem-Grid--12 aem-Grid--default--12",":items":{"experiencefragment-banner":{"id":"experiencefragment-a7cb3d4771","localizedFragmentVariationPath":"/content/experience-fragments/snowflake-site/language-masters/en/site/pushdown-banner/master/jcr:content","configured":true,":type":"snowflake-site/components/experiencefragment","xfModelPath":"/content/experience-fragments/snowflake-site/language-masters/en/site/pushdown-banner/master.xfmodel.json"},"experiencefragment-header":{"id":"experiencefragment-ab0cc5961d","localizedFragmentVariationPath":"/content/experience-fragments/snowflake-site/language-masters/en/site/mega-nav-header/master/jcr:content","configured":true,":type":"snowflake-site/components/experiencefragment","xfModelPath":"/content/experience-fragments/snowflake-site/language-masters/en/site/mega-nav-header/master.xfmodel.json","languageNavPath":"/content/snowflake-site/global/en/artificial-intelligence/industries.languagenav.json"},"responsivegrid":{"columnCount":12,"columnClassNames":{"flexible_column_cont_939100716":"aem-GridColumn aem-GridColumn--default--12","flexible_column_cont":"aem-GridColumn aem-GridColumn--default--12","flexible_column_cont_1158003461":"aem-GridColumn aem-GridColumn--default--12","flexible_column_cont_663228916":"aem-GridColumn aem-GridColumn--default--12","flexible_column_cont_912630531":"aem-GridColumn aem-GridColumn--default--12","flexible_column_cont_1398138236":"aem-GridColumn aem-GridColumn--default--12","flexible_column_cont_1467213961":"aem-GridColumn aem-GridColumn--default--12","flexible_column_cont_1377146023":"aem-GridColumn aem-GridColumn--default--12"},"gridClassNames":"aem-Grid aem-Grid--12 aem-Grid--default--12",":items":{"flexible_column_cont":{"id":"flexible-column-container-4a5d274dfe","propertiesId":"hub-hero-breadcrumbs","type":"1-column","alignColumns":"top","containerMaxWidth":"extra-large","topPadding":"small","bottomPadding":"none","spaceBetween":"small","reverseOnMobile":false,"carouselOnMobile":false,"propertiesCSSClasses":"page-section","backgroundImageOption":"none","flexible_column_content_container_1":{"layout":"SIMPLE","id":"container-bfe560eb74",":type":"snowflake-site/components/flexible-column-container/flexible-column-content-container",":items":{"breadcrumb":{"id":"breadcrumb-fe43ce5889","items":[{"id":"breadcrumb-fe43ce5889-item-e7ce355135","link":{"valid":true,"url":"/en/"},"active":false,"current":false,"title":"English",":type":"snowflake-site/components/structure/page","appliedCssClassNames":"summit-page"},{"id":"breadcrumb-fe43ce5889-item-4b75ed29e6","link":{"valid":true,"url":"/en/artificial-intelligence/"},"active":false,"current":false,"title":"Artificial Intelligence",":type":"snowflake-site/components/structure/page","appliedCssClassNames":"summit-page"},{"id":"breadcrumb-fe43ce5889-item-2ad6169dd6","link":{"valid":true,"url":"/en/artificial-intelligence/industries/"},"active":true,"current":true,"title":"AI in Industries",":type":"snowflake-site/components/structure/page","appliedCssClassNames":"summit-page"}],":type":"snowflake-site/components/breadcrumb"}},":itemsOrder":["breadcrumb"]},":type":"snowflake-site/components/flexible-column-container","isActiveTOC":false,"isBlogPage":false,"appliedCssClassNames":"snowflake-flexible-column-container-gray-10-bg"},"flexible_column_cont_939100716":{"id":"flexible-column-container-15f06486c1","propertiesId":"hub-hero","type":"2-column-even","alignColumns":"center","containerMaxWidth":"extra-large","topPadding":"extra-small","bottomPadding":"extra-small","spaceBetween":"small","reverseOnMobile":false,"carouselOnMobile":false,"propertiesCSSClasses":"page-section","backgroundImageOption":"none","flexible_column_content_container_1":{"layout":"SIMPLE","id":"container-ddaee9be59",":type":"snowflake-site/components/flexible-column-container/flexible-column-content-container",":items":{"title_v2":{"id":"title-v2-0523775201","additionalClasses":"hub-hero__headline","type":"heading1","lines":["AI in Action: A Directory of AI Use Cases Across Industries"],":type":"snowflake-site/components/title-v2","appliedCssClassNames":"left-alignment"},"text":{"id":"text-e19f362071","additionalClasses":"hub-hero__subheadline","text":"\u003Cp\u003EAI is becoming more specialized, but outcomes depend on a repeatable foundation. This guide examines how industry-specific AI applications are taking shape — and why governed data, semantic context and purpose-built infrastructure determine whether they produce reliable results.\u003C/p\u003E","richText":true,":type":"snowflake-site/components/text","appliedCssClassNames":"text-size-regular text-color-text-05"},"container":{"additionalClasses":"hub-hero__authors","layout":"RESPONSIVE_GRID","columnCount":12,"columnClassNames":{"content_chip_copy":"aem-GridColumn aem-GridColumn--default--12","content_chip":"aem-GridColumn aem-GridColumn--default--12"},"gridClassNames":"aem-Grid aem-Grid--12 aem-Grid--default--12","id":"container-687d1ce4dc",":type":"snowflake-site/components/container",":items":{"content_chip":{"id":"content-chip-81394781f9","cta":{"id":"cta","showOutboundIcon":false,"buttonLink":{"valid":true,"url":"/en/blog/authors/Laurie-Macpherson/"},"linkTargetContentType":"GENERIC",":type":"snowflake-site/components/button","linkType":"SNOWFLAKE_INTERNAL","text":"Read bio"},"image":{"id":"image","height":"800","src":"https://www.snowflake.com/adobe/dynamicmedia/deliver/dm-aid--a7e9fdff-6f08-4edc-9cd1-e213bb234aaa/laurie-macpherson.jpg?quality=85&preferwebp=true","alt":"Laurie MacPherson","lazyEnabled":true,"width":"800",":type":"snowflake-site/components/image"},"headline":{"id":"title","type":"heading5","lines":["Laurie MacPherson","Technical Writer, Snowflake"],":type":"snowflake-site/components/title-v2"},":type":"snowflake-site/components/content-chip"},"content_chip_copy":{"id":"content-chip-6fa65b2207","cta":{"id":"cta","showOutboundIcon":false,"buttonLink":{"valid":true,"url":"/en/blog/authors/david-gaule/"},"linkTargetContentType":"GENERIC",":type":"snowflake-site/components/button","linkType":"SNOWFLAKE_INTERNAL","text":"Read bio"},"image":{"id":"image","height":"512","src":"https://www.snowflake.com/adobe/dynamicmedia/deliver/dm-aid--fd9454ea-3b59-4d19-95be-534774cbd226/david.jpg?quality=85&preferwebp=true","alt":"David Gaule","lazyEnabled":true,"width":"512",":type":"snowflake-site/components/image"},"headline":{"id":"title","type":"heading5","lines":["David Gaule","Technical Editor, Snowflake"],":type":"snowflake-site/components/title-v2"},":type":"snowflake-site/components/content-chip"}},":itemsOrder":["content_chip","content_chip_copy"],"appliedCssClassNames":"snowflake-responsive-container-inner-padding-small"}},":itemsOrder":["title_v2","text","container"],"appliedCssClassNames":"snowflake-responsive-container-inner-padding-extra-small"},"flexible_column_content_container_2":{"additionalClasses":"hub-hero__video-column","layout":"SIMPLE","id":"container-c0f83aa762",":type":"snowflake-site/components/flexible-column-container/flexible-column-content-container",":items":{"youtube":{"id":"embed-9e4839b49c","youtubeVideoId":"F34xlRoQ3eQ","layout":"responsive","youtubeAspectRatio":"56.25","youtubeAutoPlay":false,"youtubeLoop":false,"youtubeMute":false,"youtubePlaysInline":false,"youtubeRel":false,"embeddableResourceType":"core/wcm/components/embed/v1/embed/embeddable/youtube","type":"EMBEDDABLE",":type":"snowflake-site/components/youtube"}},":itemsOrder":["youtube"]},":type":"snowflake-site/components/flexible-column-container","isActiveTOC":false,"isBlogPage":false,"appliedCssClassNames":"snowflake-flexible-column-container-gray-10-bg"},"flexible_column_cont_1398138236":{"id":"flexible-column-container-2c6dfd6d38","propertiesId":"hub-hero-related-topics","type":"1-column","alignColumns":"top","containerMaxWidth":"extra-large","topPadding":"extra-small","bottomPadding":"medium","spaceBetween":"small","reverseOnMobile":false,"carouselOnMobile":false,"propertiesCSSClasses":"page-section","backgroundImageOption":"none","flexible_column_content_container_1":{"additionalClasses":"related-topics-outer-container border-top","layout":"SIMPLE","id":"container-932a511c52",":type":"snowflake-site/components/flexible-column-container/flexible-column-content-container",":items":{"text_894059747":{"id":"text-546835a865","additionalClasses":"seo-hub-hero__related-topic-label","text":"\u003Cp\u003EAI in Industry Topics:\u003C/p\u003E\r\n","richText":true,":type":"snowflake-site/components/text","appliedCssClassNames":"text-size-regular text-color-text-05"},"text":{"id":"text-4cd02939db","additionalClasses":"related-topics ","text":"\u003Cul\u003E\r\n\u003Cli\u003E\u003Ca href=\"https://www.snowflake.com/en/artificial-intelligence/industries/ai-in-cybersecurity/\"\u003EAI in Cybersecurity\u003C/a\u003E\u003C/li\u003E\r\n\u003Cli\u003E\u003Ca href=\"https://www.snowflake.com/en/artificial-intelligence/industries/ai-in-retail/\"\u003EAI in Retail\u003C/a\u003E\u003C/li\u003E\r\n\u003C/ul\u003E\r\n","richText":true,":type":"snowflake-site/components/text","appliedCssClassNames":"text-size-small"}},":itemsOrder":["text_894059747","text"]},":type":"snowflake-site/components/flexible-column-container","isActiveTOC":false,"isBlogPage":false,"appliedCssClassNames":"snowflake-flexible-column-container-gray-10-bg"},"flexible_column_cont_663228916":{"id":"flexible-column-container-485ebb6644","propertiesId":"hub-body","type":"2-column-60-40","alignColumns":"top","containerMaxWidth":"extra-large","topPadding":"medium","bottomPadding":"medium","spaceBetween":"small","reverseOnMobile":false,"carouselOnMobile":false,"propertiesCSSClasses":"page-section","backgroundImageOption":"none","flexible_column_content_container_1":{"additionalClasses":"longform-content","layout":"SIMPLE","id":"hub-body-content",":type":"snowflake-site/components/flexible-column-container/flexible-column-content-container",":items":{"text_intro_0":{"id":"text-3b93c3c77c","text":"\u003Cp\u003ESpending on industry-specific generative AI applications reached an estimated $3.5 billion in 2025, nearly three times the previous year’s total, according to \u003Ca href=\"https://menlovc.com/perspective/2025-the-state-of-generative-ai-in-the-enterprise/\" target=\"_blank\"\u003EMenlo Ventures\u003C/a\u003E. This growth reflects a widening market for AI built around the workflows, data and operating requirements of particular industries.\u003C/p\u003E\n\u003Cp\u003EAcross environments, organizations are using predictive \u003Ca href=\"https://www.snowflake.com/en/artificial-intelligence/machine-learning/\"\u003Emachine learning\u003C/a\u003E, \u003Ca href=\"https://www.snowflake.com/en/artificial-intelligence/generative-ai/\"\u003Egenerative AI\u003C/a\u003E and \u003Ca href=\"https://www.snowflake.com/en/artificial-intelligence/agents/\"\u003Eagentic AI\u003C/a\u003E — often incorporating \u003Ca href=\"https://www.snowflake.com/en/artificial-intelligence/computer-vision/\"\u003Ecomputer vision\u003C/a\u003E and other multimodal capabilities — to support business processes ranging from detection and forecasting to research, decision support and operational automation.\u003C/p\u003E\n\u003Cp\u003EThis directory examines the AI use cases shaping 13 industries, how they fit into specific business processes and what organizations need to turn those capabilities into reliable results. It also identifies foundational considerations that can support successful AI initiatives.\u003C/p\u003E","richText":true,":type":"snowflake-site/components/text","appliedCssClassNames":"text-color-text-05"},"title_what-shapes-ai-outcomes-across-industries":{"id":"title-v2-8c1a01027a","additionalClasses":"anchor-title anchor-title--what-shapes-ai-outcomes-across-industries","type":"heading2","lines":["What shapes AI outcomes across industries"],":type":"snowflake-site/components/title-v2"},"text_what-shapes-ai-outcomes-across-industries_0":{"id":"text-069f2b7211","text":"\u003Cp\u003EAn AI model can provide powerful analytical and reasoning capabilities, but the surrounding system strongly influences whether the output is grounded in relevant data, interpreted through the correct business context and delivered within applicable policies and procedures. This foundation shapes how reliably the model’s capabilities can be applied inside the business.\u003C/p\u003E\n\u003Ch3\u003EGoverned, trusted data\u003C/h3\u003E\n\u003Cp\u003EAn AI system begins with the data available to it. When records are incomplete, outdated or inconsistently defined, the weakness carries into the result. An equipment-failure prediction, for example, depends on accurate sensor readings, maintenance history and asset identity. Missing service records or signals assigned to the wrong component can undermine the prediction before the model begins its analysis.\u003C/p\u003E\n\u003Cp\u003ETrust also depends on how that information is accessed and used. Sensitive data needs to remain protected as it enters model, retrieval and agent workflows. Teams need enough traceability to determine which records, models and instructions contributed to an output, particularly when the system can retrieve information, call a tool or initiate an action.\u003C/p\u003E\n\u003Cp\u003EGovernance doesn’t guarantee that a model will produce a correct result, but it does establish the conditions necessary for teams to inspect the result, identify its source and limit who or what can use the underlying information.\u003C/p\u003E\n\u003Ch3\u003ESemantic context\u003C/h3\u003E\n\u003Cp\u003EEnterprise data almost always carries meaning that’s not obvious by the label name. For example, a column labeled revenue might refer to bookings, billings or recognized revenue, each of which would produce a different answer to the same question.\u003C/p\u003E\n\u003Cp\u003EPeople working inside the organization understand these distinctions through experience, policies and established business rules. An AI system needs those things made explicit. Metric definitions, entity relationships, terminology, lineage, metadata and approved calculation logic connect a request with the correct data and help preserve the organization’s intended meaning.\u003C/p\u003E\n\u003Ch3\u003EAI infrastructure close to governed data\u003C/h3\u003E\n\u003Cp\u003EEnterprise AI applications often need structured records, documents, images, event streams and externally sourced data within the same workflow. Moving those assets into a separate AI environment can create additional copies, fragmented controls and pipelines that have to preserve permissions as data crosses system boundaries.\u003C/p\u003E\n\u003Cp\u003EKeeping model inference, retrieval, feature engineering and agent workflows close to governed data that can reduce unnecessary data movement. Depending on the architecture and configuration, teams can apply access controls, use recently updated source information and monitor data use across supported portions of the AI workflow.\u003C/p\u003E","richText":true,":type":"snowflake-site/components/text","appliedCssClassNames":"text-color-text-05"},"title_ai-use-cases-built-on-sensitive-regulated-data":{"id":"title-v2-b02407b441","additionalClasses":"anchor-title anchor-title--ai-use-cases-built-on-sensitive-regulated-data","type":"heading2","lines":["AI use cases built on sensitive, regulated data"],":type":"snowflake-site/components/title-v2"},"text_ai-use-cases-built-on-sensitive-regulated-data_0":{"id":"text-80d0a6ee10","text":"\u003Cp\u003EHealthcare, life sciences and financial services organizations work with information that is commercially valuable, personally sensitive and closely regulated. AI applications in these industries need fine-grained access controls, traceability to source records, defined human-review thresholds and monitoring across patient populations, customer segments, products and geographies.\u003C/p\u003E\r\n\u003Cp\u003ECollaboration with providers, research partners, counterparties and regulators adds another requirement: sensitive data must remain governed when analysis crosses organizational boundaries.\u003C/p\u003E\r\n\u003Ch3\u003EHealthcare and life sciences AI use cases\u003C/h3\u003E\r\n\u003Cp\u003EThese applications are designed to support — not replace — the judgment of qualified clinical or regulatory professionals. They should also be validated for the specific use case, patient population, and operating environment.\u003C/p\u003E\r\n\u003Ch4\u003EEarlier disease and risk detection\u003C/h4\u003E\r\n\u003Cp\u003EPredictive models can combine laboratory results, diagnoses, medication history, clinical observations and prior utilization to identify patients whose risk is rising before a single result would trigger concern. To produce operational value, the score must connect to a defined response — such as clinician review, additional testing or care management — and be evaluated through measures including avoidable admissions, time to diagnosis and complication rates.\u003C/p\u003E\r\n\u003Ch4\u003EClinical decision support\u003C/h4\u003E\r\n\u003Cp\u003EAI can retrieve relevant evidence, summarize a patient’s recent history and compare treatment options with contraindications or risk factors. Because incomplete summaries can be as consequential as incorrect ones, clinicians need access to the underlying record, clear indications of uncertainty and authority over the final decision.\u003C/p\u003E\r\n\u003Ch4\u003EMedical imaging and diagnostic assistance\u003C/h4\u003E\r\n\u003Cp\u003EComputer vision can help identify findings in radiology images, pathology slides, retinal scans and other diagnostic media, while multimodal systems combine those images with reports, laboratory results and patient history. Validation needs to account for scanner type, image quality, acquisition protocol and patient population, with results measured through reading time, diagnostic turnaround, screening throughput or detection rates.\u003C/p\u003E\r\n\u003Ch4\u003EPatient flow and capacity planning\u003C/h4\u003E\r\n\u003Cp\u003EForecasting models can estimate emergency department arrivals, inpatient demand, discharge timing and procedural volume, while optimization methods allocate beds, operating rooms, infusion chairs and staff. The model must represent scheduled procedures, patient acuity, staffing constraints and downstream capacity; useful measures include wait time, canceled procedures, overtime and length of stay.\u003C/p\u003E\r\n\u003Ch4\u003ERevenue cycle and claims operations\u003C/h4\u003E\r\n\u003Cp\u003EAI can extract information from clinical documentation, suggest coding, identify missing records and predict denials. Agents can gather supporting documents, check payer requirements and prepare appeals for specialist review, provided the workflow preserves links among clinical events, authorizations, contracts, codes and claim status. Denial rates, days in accounts receivable and manual touches per claim show whether the application reduces rework.\u003C/p\u003E\r\n\u003Ch4\u003EPatient engagement and care navigation\u003C/h4\u003E\r\n\u003Cp\u003EAI-assisted navigation can answer preparation or coverage questions, direct patients to appropriate resources and tailor reminders to a care plan. Predictive models can also identify patients at risk of missing appointments or discontinuing treatment, with personalization shaped by consent, channel preference, language and health literacy. Appointment completion, care-gap closure and contact-center volume provide practical measures.\u003C/p\u003E\r\n\u003Ch4\u003EDrug discovery and clinical development\u003C/h4\u003E\r\n\u003Cp\u003EMachine learning can help prioritize therapeutic targets, predict molecular properties and narrow candidate compounds before laboratory validation. During clinical development, AI can support protocol design, site selection, patient matching, safety-signal review and trial forecasting using scientific literature, assay data, molecular structures, trial records and real-world evidence. Because these outputs inform experimental and regulatory judgment, reproducibility and version control remain essential.\u003C/p\u003E\r\n\u003Ch4\u003EPharmacovigilance and regulatory intelligence\u003C/h4\u003E\r\n\u003Cp\u003EDocument intelligence can extract adverse events from case reports, medical literature and call notes, while classification models route cases by seriousness, expectedness and reporting deadline. Generative AI can summarize cases or compare new guidance with existing obligations, provided the system preserves reliable entity resolution and links every conclusion to the source evidence.\u003C/p\u003E\r\n","richText":true,":type":"snowflake-site/components/text","appliedCssClassNames":"text-color-text-05"},"card_v2_ai-use-cases-built-on-sensitive-regulated-data_0":{"id":"card-v2-13ccb4f5e7","additionalClasses":"seo-customer","configurationStatus":{"configured":true,"message":""},":type":"snowflake-site/components/card-v2","text":{"id":"text","text":"\u003Cp\u003ESydney Local Health District uses the Snowflake AI Data Cloud to give its Women and Babies Service near real-time visibility into maternity data, helping clinicians act faster on births involving medical intervention, trauma, mortality and morbidity. With Snowflake, the team reduced data query times from more than one hour to 55 seconds, expanded maternity data retrieval from three months to several years, and can provision new reporting requests in hours instead of months.\u003C/p\u003E\r\n","richText":true,":type":"snowflake-site/components/text"},"layoutStyle":"horizontal","title":{"id":"title","type":"heading4","lines":["Customer story: Sydney Local Health District"],":type":"snowflake-site/components/title-v2"},"button":{"id":"button","showOutboundIcon":false,"buttonLink":{"valid":true,"url":"/en/customers/all-customers/case-study/sydney-local-health-district/"},"linkTargetContentType":"DOCUMENT",":type":"snowflake-site/components/button","linkType":"SNOWFLAKE_INTERNAL","text":"Read the full case study"},"image":{"id":"image","height":"351","src":"https://www.snowflake.com/adobe/dynamicmedia/deliver/dm-aid--f87e3a37-db0d-48e3-bfbd-6770e15a8be9/sydney-local-health-district%25403x.png?quality=85&preferwebp=true","alt":"Sydney Local Health District","lazyEnabled":true,"width":"624",":type":"snowflake-site/components/image"},"type":"content-card"},"text_ai-use-cases-built-on-sensitive-regulated-data_1":{"id":"text-7d30b62500","text":"\u003Ch3\u003EFinancial services AI use cases\u003C/h3\u003E\r\n\u003Ch4\u003EFraud detection and transaction monitoring\u003C/h4\u003E\r\n\u003Cp\u003EFraud models evaluate device, location, merchant, account history, payment pattern and network relationships to identify suspicious activity. Graph techniques can reveal coordinated behavior across accounts that appear legitimate individually, while detection rate, false positives, investigation time and disruption to legitimate customers indicate operational performance.\u003C/p\u003E\r\n\u003Ch4\u003EAnti-money laundering investigations\u003C/h4\u003E\r\n\u003Cp\u003EAI can prioritize alerts, connect related entities and summarize evidence across know-your-customer records, transaction history, prior cases and external risk sources. Investigators need to reconstruct how the system identified each relationship, particularly when the output contributes to a regulatory filing.\u003C/p\u003E\r\n\u003Ch4\u003ECredit risk and underwriting\u003C/h4\u003E\r\n\u003Cp\u003EPredictive models can estimate default risk, loss severity and affordability from application data, account behavior and cash-flow information, while document intelligence extracts financial details from statements, tax records and contracts. Institutions need clear reasons for decisions, manual-review paths and monitoring across borrower segments, products and economic conditions.\u003C/p\u003E\r\n\u003Cp\u003EUse in credit decisions must also comply with applicable fair-lending, adverse-action, explainability and consumer-protection requirements.\u003C/p\u003E\r\n\u003Ch4\u003EPersonalized financial experiences\u003C/h4\u003E\r\n\u003Cp\u003EA customer 360 can connect account activity, product holdings, transactions and interaction history so institutions can recommend a relevant next action. Eligibility, suitability and consent must shape the result alongside predicted interest, with product adoption, retention and service demand providing a fuller measure than engagement alone.\u003C/p\u003E\r\n\u003Ch4\u003EMarket, liquidity and portfolio risk\u003C/h4\u003E\r\n\u003Cp\u003EIn investment banking, machine learning can help monitor changing exposures, flag unusual correlations, and enhance scenario analysis across positions and counterparties. Natural-language interfaces help analysts explore risk drivers, while generative AI summarizes changes across portfolios or reporting periods. These applications depend on consistent identifiers, pricing data, position history and approved calculations.\u003C/p\u003E\r\n\u003Ch4\u003ERegulatory compliance and reporting\u003C/h4\u003E\r\n\u003Cp\u003EDocument intelligence can compare new regulatory text with existing policies, controls and reporting obligations, while agents gather evidence and prepare portions of recurring submissions. The knowledge base must preserve jurisdiction, effective date and exceptions, with legal or compliance review retained for novel interpretations.\u003C/p\u003E\r\n\u003Ch4\u003EOperations automation\u003C/h4\u003E\r\n\u003Cp\u003EAI can classify requests, extract information from forms, reconcile records and route exceptions. Agents may gather documents for an account change, validate them against policy and prepare the update for approval. Straight-through processing, rework and exception backlog show whether automation is improving the process rather than hiding its complexity.\u003C/p\u003E\r\n\u003Ch4\u003EInsurance pricing, claims and loss prevention\u003C/h4\u003E\r\n\u003Cp\u003EInsurers can use AI for risk estimation, claims triage, damage assessment and fraud detection. Computer vision evaluates images of damaged property or vehicles, while generative AI summarizes adjuster notes and policy language. Accurate relationships among policyholders, covered assets, incidents and prior claims support measures such as claims cycle time, leakage, fraud losses and underwriting consistency.\u003C/p\u003E\r\n","richText":true,":type":"snowflake-site/components/text","appliedCssClassNames":"text-color-text-05"},"title_ai-use-cases-built-on-machine-and-operational-data":{"id":"title-v2-b6a13954e8","additionalClasses":"anchor-title anchor-title--ai-use-cases-built-on-machine-and-operational-data","type":"heading2","lines":["AI use cases built on machine and operational data"],":type":"snowflake-site/components/title-v2"},"text_ai-use-cases-built-on-machine-and-operational-data_0":{"id":"text-eda1f5915e","text":"\u003Cp\u003EManufacturing, automotive, logistics, energy and construction organizations generate continuous records of how physical assets behave. AI applications in these environments depend on reliable telemetry ingestion, synchronized time stamps, asset hierarchies, and maintenance or failure history that gives each signal meaning. Some decisions also require edge processing, although inspection results and operating events still need to return to a governed environment for broader analysis.\u003C/p\u003E\r\n\u003Cp\u003EFor safety-critical or high-consequence applications, AI outputs require use-case-specific validation, defined operating limits, appropriate human oversight and compliance with applicable industry requirements.\u003C/p\u003E\r\n\u003Ch3\u003EManufacturing AI use cases\u003C/h3\u003E\r\n\u003Ch4\u003EPredictive maintenance\u003C/h4\u003E\r\n\u003Cp\u003EPredictive maintenance models combine vibration, temperature, pressure, acoustic signals and service history to estimate failure risk or remaining useful life. Alerts must arrive early enough to affect labor, parts and production planning without creating excessive false alarms, so manufacturers typically evaluate downtime, emergency work orders, maintenance cost and asset availability.\u003C/p\u003E\r\n\u003Ch4\u003EVisual quality inspection\u003C/h4\u003E\r\n\u003Cp\u003EComputer vision can identify scratches, contamination, dimensional defects, missing components and assembly errors at production speed. Reliable inspection depends on consistent lighting, camera placement and image capture, along with human review for uncertain cases and a process for adding newly discovered defect types.\u003C/p\u003E\r\n\u003Ch4\u003EProcess optimization\u003C/h4\u003E\r\n\u003Cp\u003EAI can model the relationship among machine settings, material properties, environmental conditions and finished-product quality, then recommend adjustments that reduce scrap, cycle time or energy use. Many manufacturers begin with decision support and automate only adjustments that remain reliable within defined operating ranges.\u003C/p\u003E\r\n\u003Ch4\u003EProduction planning and demand alignment\u003C/h4\u003E\r\n\u003Cp\u003EDemand forecasts can feed production plans that account for inventory, supplier lead times, changeover costs and available capacity. Product hierarchies, substitution rules and shared components determine whether the plan is feasible, with shortages, excess inventory and schedule disruption providing useful outcome measures.\u003C/p\u003E\r\n\u003Ch4\u003EDigital twins and simulation\u003C/h4\u003E\r\n\u003Cp\u003EA digital twin — a virtual model of a physical asset — combines design information, asset configuration, sensor readings and operating history in a continuously updated representation. Teams can simulate process changes, maintenance strategies or production schedules without disrupting live operations, provided the twin remains aligned with equipment modifications and current operating conditions.\u003C/p\u003E\r\n\u003Ch4\u003ESupply chain risk and inventory optimization\u003C/h4\u003E\r\n\u003Cp\u003EAI can estimate supplier delays, forecast component demand and identify dependencies that create production risk. Agents can monitor purchase orders, inventory, supplier communications and external disruptions, then present planners with affected products and response options. Broader visibility across supplier tiers improves the analysis when collaboration agreements permit it.\u003C/p\u003E\r\n\u003Ch4\u003EEngineering and maintenance knowledge retrieval\u003C/h4\u003E\r\n\u003Cp\u003EA generative AI assistant can search manuals, engineering drawings, standard operating procedures and maintenance records, then surface the relevant procedure, recent repairs and compatible parts. Accurate asset identity and document versioning are essential, particularly when safety or warranty requirements apply.\u003C/p\u003E\r\n\u003Ch3\u003EAutomotive AI use cases\u003C/h3\u003E\r\n\u003Ch4\u003EAdvanced driver-assistance systems\u003C/h4\u003E\r\n\u003Cp\u003EAdvanced driver-assistance systems combine computer vision, radar, lidar and vehicle telemetry to support lane keeping, object detection and emergency braking. Development requires diverse driving data and validation across road design, weather, geography, lighting, vehicle configuration and rare scenarios, with traceability from each model version to its test data.\u003C/p\u003E\r\n\u003Ch4\u003EConnected-vehicle diagnostics\u003C/h4\u003E\r\n\u003Cp\u003EAI can analyze vehicle telemetry to detect developing faults and identify recurring issues across a fleet when connected with repair and warranty records. Because the same reading may mean different things across model years, component suppliers or software versions, reliable diagnostics depend on precise vehicle configuration and service history.\u003C/p\u003E\r\n\u003Ch4\u003EBattery health and range prediction\u003C/h4\u003E\r\n\u003Cp\u003EFor electric vehicles, machine learning can estimate battery state of health and remaining range from charge cycles, temperature, driving behavior and cell condition. Models need to remain reliable during temperature extremes, rapid charging and atypical driving patterns, where inaccurate estimates carry the greatest operational cost.\u003C/p\u003E\r\n\u003Ch4\u003EManufacturing quality and warranty analytics\u003C/h4\u003E\r\n\u003Cp\u003EAI can connect factory measurements, supplier batches, vehicle configurations and warranty claims to identify recurring defects. Language models extract patterns from technician narratives, while computer vision supports production inspection. Detailed bill-of-materials and software histories help narrow corrective action to the affected vehicles.\u003C/p\u003E\r\n\u003Ch4\u003ESoftware-defined vehicle development\u003C/h4\u003E\r\n\u003Cp\u003EAI can analyze test results, logs and defect reports across many hardware and software combinations. \u003Ca href=\"https://www.snowflake.com/en/artificial-intelligence/agents/coding-agents/\"\u003ECoding agents\u003C/a\u003E assist with bounded development tasks, while anomaly detection surfaces regressions across test fleets. Strong configuration control is necessary because a defect may appear only in one hardware revision, model year or software build.\u003C/p\u003E\r\n\u003Ch4\u003EPersonalized in-vehicle experiences\u003C/h4\u003E\r\n\u003Cp\u003ERecommendation systems can tailor navigation, entertainment, climate and charging suggestions to the current driver and context, while natural-language interfaces simplify access to vehicle functions. Privacy controls must distinguish vehicle-level data from individual driver data, and recommendations must account for distraction and safety.\u003C/p\u003E\r\n\u003Ch4\u003EDealer, service and parts operations\u003C/h4\u003E\r\n\u003Cp\u003EDemand forecasting can help dealers anticipate service volume and parts requirements, while agents coordinate scheduling, repair authorization and customer updates. Current inventory, technician capacity and vehicle history support measures such as first-time fix rate, appointment lead time and parts availability.\u003C/p\u003E\r\n\u003Ch3\u003ELogistics AI use cases\u003C/h3\u003E\r\n\u003Ch4\u003ERoute and network optimization\u003C/h4\u003E\r\n\u003Cp\u003ERoute optimization models balance delivery windows, vehicle capacity, traffic, fuel cost, driver schedules and depot constraints across a network. Plans can update as conditions change, with on-time delivery, empty miles, emissions and cost per stop providing practical measures.\u003C/p\u003E\r\n\u003Ch4\u003EWarehouse and fulfillment optimization\u003C/h4\u003E\r\n\u003Cp\u003EAI can forecast labor demand, select storage locations, optimize picking paths and detect inventory discrepancies. In highly automated facilities, models may coordinate equipment or surface bottlenecks. Each site needs local operating context because warehouse topology, service requirements and order profiles differ.\u003C/p\u003E\r\n\u003Ch4\u003EFleet maintenance and utilization\u003C/h4\u003E\r\n\u003Cp\u003ETelematics, inspections and repair history can support predictive maintenance and reveal idle or poorly positioned assets across trucks, trailers, aircraft, ships and material-handling equipment. Reliable asset identity is critical when tractors, trailers and components move independently.\u003C/p\u003E\r\n\u003Ch4\u003EShipment risk and exception management\u003C/h4\u003E\r\n\u003Cp\u003EAgents can monitor weather, location events, customs status and carrier communications to identify shipments at risk, then assemble alternative routes, inventory options and customer commitments for review. Approval boundaries should reflect the consequence of each action, from recommending a route to changing a delivery promise or carrier.\u003C/p\u003E\r\n\u003Ch4\u003EDocument and customs automation\u003C/h4\u003E\r\n\u003Cp\u003EDocument intelligence can extract and compare fields across bills of lading, packing lists, invoices and customs forms, then route missing or inconsistent information to specialists. Agents may also assemble documentation packages and track completion, with clearance time, storage charges and manual touches showing whether the workflow improves.\u003C/p\u003E\r\n\u003Ch4\u003EDemand and capacity forecasting\u003C/h4\u003E\r\n\u003Cp\u003EForecasting models can estimate shipment volume by lane, customer and service level, helping carriers position equipment and negotiate capacity. Because greater demand in one lane can create shortages elsewhere, the model must support networkwide planning rather than local optimization alone.\u003C/p\u003E\r\n\u003Ch3\u003EConstruction AI use cases\u003C/h3\u003E\r\n\u003Ch4\u003EProject schedule and delay prediction\u003C/h4\u003E\r\n\u003Cp\u003EAI can combine project plans, daily logs, procurement status, workforce data and historical performance to identify activities likely to affect the critical path. A consistent project hierarchy across schedules, contracts and field systems supports measures such as schedule variance, recovery time and risks identified before milestone impact.\u003C/p\u003E\r\n\u003Ch4\u003ECost forecasting and change-order risk\u003C/h4\u003E\r\n\u003Cp\u003EModels can forecast cost at completion, identify unusual spending and surface scope changes likely to produce overruns or disputes. Document intelligence connects change requests with contract terms, drawings, approvals and correspondence, while version control ensures the system uses the language currently governing the work.\u003C/p\u003E\r\n\u003Ch4\u003ESite safety monitoring\u003C/h4\u003E\r\n\u003Cp\u003EComputer vision can identify missing protective equipment, unsafe proximity to machinery and entry into restricted areas, while predictive models combine incident history, weather, schedule pressure and workforce patterns to identify elevated risk. Calibration, alert severity and clear ownership determine whether the system improves response.\u003C/p\u003E\r\n\u003Ch4\u003EProgress verification\u003C/h4\u003E\r\n\u003Cp\u003EImages from cameras, drones and mobile devices can be compared with building information models and project schedules to estimate installed quantities or construction progress. Consistent capture protocols help the model distinguish completed work from staged materials or partially installed components.\u003C/p\u003E\r\n\u003Ch4\u003EEquipment and fleet optimization\u003C/h4\u003E\r\n\u003Cp\u003ETelematics can support maintenance, utilization analysis and allocation of cranes, excavators and other equipment across sites. Idle time, rental expense, unplanned downtime and transport between locations indicate whether the application is improving asset use.\u003C/p\u003E\r\n\u003Ch4\u003EContract and document intelligence\u003C/h4\u003E\r\n\u003Cp\u003EGenerative AI can search contracts, specifications, requests for information, submittals and meeting records to answer project questions or identify conflicting requirements. The system must retrieve the current approved version while retaining superseded records for audit and dispute resolution.\u003C/p\u003E\r\n\u003Ch4\u003EBid and resource planning\u003C/h4\u003E\r\n\u003Cp\u003EAI can estimate cost, labor demand and material requirements using design documents, location, historical bids and market conditions. Because cost codes, delivery methods and regional assumptions vary, historical data needs normalization before it can support reliable comparisons.\u003C/p\u003E\r\n","richText":true,":type":"snowflake-site/components/text","appliedCssClassNames":"text-color-text-05"},"card_v2_ai-use-cases-built-on-machine-and-operational-data_0":{"id":"card-v2-f3e34b48e8","additionalClasses":"seo-customer","configurationStatus":{"configured":true,"message":""},":type":"snowflake-site/components/card-v2","text":{"id":"text","text":"\u003Cp\u003ESiemens Energy uses Snowflake and Cortex AI to transform paper-based technical information into searchable, actionable data for employees across the business. With Snowflake’s AI Data Cloud, Cortex AI and a Streamlit-powered interface, Siemens Energy made more than 800,000 pages accessible through AI chatbots, helping teams find relevant information for design optimization, cost changes, training and future use cases such as sales documents and factory procedures — while keeping data secure inside Snowflake.\u003C/p\u003E\r\n","richText":true,":type":"snowflake-site/components/text"},"layoutStyle":"horizontal","title":{"id":"title","type":"heading4","lines":["Customer story: Siemens Energy"],":type":"snowflake-site/components/title-v2"},"button":{"id":"button","showOutboundIcon":false,"buttonLink":{"valid":true,"url":"/en/customers/all-customers/video/siemens-energy/"},"linkTargetContentType":"GENERIC",":type":"snowflake-site/components/button","linkType":"SNOWFLAKE_INTERNAL","text":"Watch the customer story"},"image":{"id":"image","height":"234","src":"https://www.snowflake.com/adobe/dynamicmedia/deliver/dm-aid--9d8b3fc8-6209-4a5e-a714-287c2c1b1f76/siemens-energy.png?quality=85&preferwebp=true","alt":"Siemens Energy Logo","lazyEnabled":true,"width":"416",":type":"snowflake-site/components/image"},"type":"content-card"},"title_ai-use-cases-built-on-documents-and-language":{"id":"title-v2-7ec4738dcf","additionalClasses":"anchor-title anchor-title--ai-use-cases-built-on-documents-and-language","type":"heading2","lines":["AI use cases built on documents and language"],":type":"snowflake-site/components/title-v2"},"text_ai-use-cases-built-on-documents-and-language_0":{"id":"text-dab04fa03c","text":"\u003Cp\u003ELegal services and education store much of their operational knowledge in contracts, policies, research, instructional materials and other unstructured data. AI applications in these environments need reliable parsing across scans, PDFs, tables and forms; permission-aware retrieval; document versioning; citations; and clear separation between source-grounded output and model inference.\u003C/p\u003E\r\n\u003Ch3\u003ELegal AI use cases\u003C/h3\u003E\r\n\u003Ch4\u003EContract analysis and review\u003C/h4\u003E\r\n\u003Cp\u003EDocument intelligence can extract parties, dates, obligations, renewal terms and liability provisions, while generative AI compares language with an approved playbook and summarizes deviations. The system must recognize meaning across varied wording, with cycle time, review consistency and material issues surfaced before signature providing useful measures.\u003C/p\u003E\r\n\u003Ch4\u003EContract portfolio intelligence\u003C/h4\u003E\r\n\u003Cp\u003EAI can identify agreements affected by a regulatory change, pricing event, acquisition or supplier disruption across an existing portfolio. Accurate entity resolution, amendment chains and effective dates determine which counterparty and obligation currently apply.\u003C/p\u003E\r\n\u003Ch4\u003ELegal research and precedent retrieval\u003C/h4\u003E\r\n\u003Cp\u003EGenerative AI can retrieve cases, statutes, regulations, internal memoranda and prior work product, then organize the authorities associated with a question. Professional use requires verifiable citations, recognition of superseded material and distinction between controlling and persuasive authority.\u003C/p\u003E\r\n\u003Ch4\u003ELitigation and investigation support\u003C/h4\u003E\r\n\u003Cp\u003EAI can classify documents, identify entities, reconstruct timelines and surface conceptually related communications across large review sets. Legal teams need reproducibility, chain of custody and preserved records of search criteria, model versions and review decisions.\u003C/p\u003E\r\n\u003Ch4\u003ERegulatory change management\u003C/h4\u003E\r\n\u003Cp\u003EDocument intelligence can compare new regulatory text with policies and controls, while agents prepare an initial impact assessment and route follow-up work. Jurisdiction, effective date and exceptions must remain attached to the analysis, with legal review retained for interpretation.\u003C/p\u003E\r\n\u003Ch4\u003EMatter management and legal operations\u003C/h4\u003E\r\n\u003Cp\u003EAI can classify incoming requests, route work by expertise and capacity, forecast matter cost and identify billing anomalies. Consistent matter coding and billing data support measures such as intake time, budget variance, outside-counsel spend and workload visibility.\u003C/p\u003E\r\n\u003Ch4\u003EKnowledge management and drafting support\u003C/h4\u003E\r\n\u003Cp\u003EA permission-aware assistant can retrieve approved templates, clauses, guidance and prior work product before preparing a first draft. Clear ownership and lifecycle controls keep outdated templates or superseded guidance from entering current work.\u003C/p\u003E\r\n\u003Cp\u003EAI-generated legal analysis should be reviewed by a qualified professional before it’s used for legal advice, filings or other important decisions.\u003C/p\u003E\r\n\u003Ch3\u003EEducation AI use cases\u003C/h3\u003E\r\n\u003Ch4\u003EIntelligent tutoring and learner support\u003C/h4\u003E\r\n\u003Cp\u003EIntelligent tutoring systems can adapt explanations, practice questions and feedback to a learner’s progress using curriculum content and prior performance. Institutions need alignment with approved learning objectives and should evaluate mastery and progression rather than engagement alone.\u003C/p\u003E\r\n\u003Ch4\u003EStudent success and retention analytics\u003C/h4\u003E\r\n\u003Cp\u003EPredictive models can combine attendance, course activity, assessment performance and advising history to identify students who may need support. A useful score connects to a defined intervention, includes understandable contributing factors and is monitored across student groups, with retention, completion and response to outreach providing practical measures.\u003C/p\u003E\r\n\u003Ch4\u003EAutomated grading and feedback\u003C/h4\u003E\r\n\u003Cp\u003EAI can score objective work, compare responses with a rubric and prepare formative feedback for instructor review. Open-ended work requires representative examples, rubric alignment and escalation for ambiguous cases, particularly when the output affects final grades or progression. Organizations should retain appropriate educator oversight for consequential grading, placement, progression and student-support decisions.\u003C/p\u003E\r\n\u003Ch4\u003EInstitutional knowledge and policy assistance\u003C/h4\u003E\r\n\u003Cp\u003EA natural-language assistant can answer questions using handbooks, policies and program requirements. Because the correct answer may depend on location, enrollment year, program or employee type, the system needs version control, audience-specific access and citations to the applicable policy.\u003C/p\u003E\r\n\u003Ch4\u003EResearch administration and grant operations\u003C/h4\u003E\r\n\u003Cp\u003EAgents can classify funding opportunities, compare requirements, assemble compliance documents and track reporting deadlines, while document intelligence extracts terms from award materials. Proposal preparation time, missed deadlines and administrative workload provide practical measures.\u003C/p\u003E\r\n","richText":true,":type":"snowflake-site/components/text","appliedCssClassNames":"text-color-text-05"},"title_ai-use-cases-built-on-behavioral-and-event-data":{"id":"title-v2-356c545e17","additionalClasses":"anchor-title anchor-title--ai-use-cases-built-on-behavioral-and-event-data","type":"heading2","lines":["AI use cases built on behavioral and event data"],":type":"snowflake-site/components/title-v2"},"text_ai-use-cases-built-on-behavioral-and-event-data_0":{"id":"text-833bfeba60","text":"\u003Cp\u003ERetail, advertising, media and cybersecurity organizations rely on high-volume records of what customers, audiences, systems and devices do over time. These applications require identity resolution across devices and channels, event processing matched to the decision window, consent and purpose controls, and feedback loops that distinguish a click, alert or anomaly from the outcome the organization actually cares about.\u003C/p\u003E\r\n\u003Ch3\u003ERetail and ecommerce AI use cases\u003C/h3\u003E\r\n\u003Ch4\u003EDemand forecasting\u003C/h4\u003E\r\n\u003Cp\u003EDemand forecasting models combine transaction history, pricing, promotions, seasonality, inventory and external signals to estimate demand by product, location and channel. Product and location hierarchies must reflect substitutions, assortment changes and new products, with availability, markdowns, spoilage and inventory turns providing practical measures.\u003C/p\u003E\r\n\u003Ch4\u003EAssortment and inventory optimization\u003C/h4\u003E\r\n\u003Cp\u003EOptimization models translate demand forecasts into stocking and allocation decisions under shelf, supplier, margin and capacity constraints. They also need to represent interactions across the assortment, since adding one product or promotion may shift demand elsewhere.\u003C/p\u003E\r\n\u003Ch4\u003EPersonalization and recommendation engines\u003C/h4\u003E\r\n\u003Cp\u003ERecommendation engines rank products or content using browsing, purchase, search and product data, while a customer 360 supports consistency across ecommerce, stores, service and marketing. Availability, margin, consent, merchandising priorities and customer fatigue should shape the result alongside predicted interest.\u003C/p\u003E\r\n\u003Ch4\u003EPricing and promotion optimization\u003C/h4\u003E\r\n\u003Cp\u003EAI can estimate demand response, promotion cannibalization and markdown timing, allowing merchants to compare revenue, margin and inventory scenarios. Recommendations require ongoing recalibration because competitor behavior, supply constraints and customer sentiment can change historical elasticity.\u003C/p\u003E\r\n\u003Ch4\u003ERetail media optimization\u003C/h4\u003E\r\n\u003Cp\u003ERetail media networks can use first-party shopping behavior to build audiences, forecast performance and connect ad exposure with sales, while \u003Ca href=\"https://www.snowflake.com/en/fundamentals/what-is-a-data-clean-room/\"\u003Edata clean rooms\u003C/a\u003E support collaboration with brands without exposing raw customer records. Agreed definitions for exposure, conversion and incrementality are necessary for reliable measurement.\u003C/p\u003E\r\n\u003Ch4\u003EStore operations and cashier-less checkout\u003C/h4\u003E\r\n\u003Cp\u003ERetail applications of computer vision and sensor fusion can support cashier-less checkout, shelf monitoring, queue detection and loss prevention, while forecasting aligns labor with traffic and fulfillment demand. Store layout, lighting and customer behavior require local calibration and clear exception handling.\u003C/p\u003E\r\n\u003Ch4\u003ESupply chain and fulfillment orchestration\u003C/h4\u003E\r\n\u003Cp\u003EAI can predict supplier delays, select fulfillment locations and recommend delivery options based on inventory, promised date and cost. Current inventory is critical, since stale data can create split shipments, cancellations or missed promises elsewhere in the network.\u003C/p\u003E\r\n\u003Ch4\u003EProduct content and commerce automation\u003C/h4\u003E\r\n\u003Cp\u003EGenerative AI can extract attributes from supplier files, create descriptions, translate content and identify incomplete catalog records, while multimodal models compare product images with listed attributes. Human review should focus on regulated claims, brand language and high-value products.\u003C/p\u003E\r\n\u003Cp\u003E\u003Ca href=\"https://www.snowflake.com/en/artificial-intelligence/industries/ai-in-retail/\"\u003ELearn more about AI in retail >\u003C/a\u003E\u003C/p\u003E\r\n\u003Ch3\u003EAdvertising AI use cases\u003C/h3\u003E\r\n\u003Ch4\u003EAudience creation and expansion\u003C/h4\u003E\r\n\u003Cp\u003EAI can identify audience segments from behavioral data, purchase signals, content interaction and campaign response, then extend them with lookalike models. The outcome chosen to define value — clicks, qualified leads, purchases or long-term customer value — shapes the resulting audience.\u003C/p\u003E\r\n\u003Ch4\u003EMedia planning and budget allocation\u003C/h4\u003E\r\n\u003Cp\u003EPredictive models can forecast reach, conversion and marginal return across channels, helping advertisers compare additional spend. Because channel data uses different identifiers, attribution windows and reporting methods, consistent measurement is required before performance can be compared meaningfully.\u003C/p\u003E\r\n\u003Ch4\u003EBidding and campaign optimization\u003C/h4\u003E\r\n\u003Cp\u003EAI can adjust bids, placements and pacing according to predicted response and campaign goals. Spend limits, inventory quality and prohibited placements remain explicit controls, while incremental conversion and profit provide stronger measures than inexpensive impressions alone.\u003C/p\u003E\r\n\u003Ch4\u003ECreative intelligence and generation\u003C/h4\u003E\r\n\u003Cp\u003EMultimodal AI can classify creative attributes, connect them with performance and generate variations for different audiences or formats. Brand, legal and rights controls remain necessary, and experimentation should isolate the effect of the creative from other campaign changes.\u003C/p\u003E\r\n\u003Ch4\u003EMeasurement and attribution\u003C/h4\u003E\r\n\u003Cp\u003EAI can estimate campaign contribution across fragmented customer journeys, while data clean rooms allow advertisers, publishers and retailers to compare exposure and outcome data under governed conditions. Holdouts, incrementality tests and clearly stated assumptions help executives interpret attribution without overstating precision.\u003C/p\u003E\r\n\u003Ch4\u003EFraud and inventory quality\u003C/h4\u003E\r\n\u003Cp\u003EAnomaly detection can identify invalid traffic, bot activity, domain spoofing and suspicious conversion patterns, while graph analysis surfaces related sites, devices or accounts. Current event data and analyst feedback are necessary because fraud behavior changes in response to detection.\u003C/p\u003E\r\n\u003Ch4\u003EPrivacy-aware collaboration\u003C/h4\u003E\r\n\u003Cp\u003EData clean rooms can enable organizations to match and analyze data while limiting direct exposure of raw customer records. Identity rules, aggregation thresholds and permitted uses must be defined before analysis; the technical environment can help enforce configured rules, but organizations remain responsible for defining and governing permitted uses.\u003C/p\u003E\r\n\u003Ch3\u003EMedia and entertainment AI use cases\u003C/h3\u003E\r\n\u003Ch4\u003EContent recommendation and discovery\u003C/h4\u003E\r\n\u003Cp\u003ERecommendation engines rank programs, articles, music or games using audience behavior, session context and catalog metadata. Accurate information about genre, cast, topic, rights and format helps balance immediate relevance with discovery, diversity and long-term engagement.\u003C/p\u003E\r\n\u003Ch4\u003EAudience analytics and churn prediction\u003C/h4\u003E\r\n\u003Cp\u003EPredictive models can identify engagement patterns associated with cancellation, upgrades or reduced use, allowing media companies to tailor retention offers or content recommendations. The intervention should be tested to determine whether it changes behavior rather than assuming every at-risk subscriber can be retained.\u003C/p\u003E\r\n\u003Ch4\u003EContent demand and commissioning\u003C/h4\u003E\r\n\u003Cp\u003EAI can analyze audience trends, catalog performance, search behavior and market signals to inform acquisition, programming and commissioning. Historical performance may favor familiar concepts, so the output should support portfolio analysis while leaving editorial and strategic judgment intact.\u003C/p\u003E\r\n\u003Ch4\u003EAdvertising yield and inventory optimization\u003C/h4\u003E\r\n\u003Cp\u003EModels can forecast available inventory, audience composition and demand, then help allocate impressions among direct commitments, programmatic demand and internal promotion. Rights, frequency and campaign constraints must remain current, with revenue per impression, fill rate and delivery against commitments providing useful measures.\u003C/p\u003E\r\n\u003Ch4\u003EContent metadata and archive intelligence\u003C/h4\u003E\r\n\u003Cp\u003EMultimodal AI can transcribe, classify and tag video, audio and images, making archives searchable by topic, person, scene or spoken phrase. Rights information must remain connected to each asset so improved discovery doesn’t lead to unauthorized use.\u003C/p\u003E\r\n\u003Ch4\u003ELocalization and accessibility\u003C/h4\u003E\r\n\u003Cp\u003EAI can assist with transcription, translation, captioning, dubbing and audio description, while specialists focus on cultural adaptation, timing and quality. Evaluation should account for language, dialect, genre, background audio, accessibility and audience experience.\u003C/p\u003E\r\n\u003Ch4\u003EProduction and postproduction workflows\u003C/h4\u003E\r\n\u003Cp\u003EAI can organize footage, search transcripts, identify continuity issues and generate rough edits or visual effects assets, while agents coordinate review status and versions. Unreleased content requires strict access and rights controls, with early value often coming from faster logging, retrieval and coordination.\u003C/p\u003E\r\n\u003Ch4\u003ERights, royalties and content protection\u003C/h4\u003E\r\n\u003Cp\u003EAI can match content usage with ownership records, identify unlicensed distribution and detect anomalies in royalty reporting. Because rights vary by territory, platform and time period, accurate relationships among assets, contracts and owners are necessary before the system can support payment or enforcement decisions.\u003C/p\u003E\r\n","richText":true,":type":"snowflake-site/components/text","appliedCssClassNames":"text-color-text-05"},"card_v2_ai-use-cases-built-on-behavioral-and-event-data_0":{"id":"card-v2-9111d8a683","additionalClasses":"seo-customer","configurationStatus":{"configured":true,"message":""},":type":"snowflake-site/components/card-v2","text":{"id":"text","text":"\u003Cp\u003ELuminate uses Snowflake as the core of its entertainment data platform to unify more than 3.5 TB of daily data across music, film and television, powering products such as the Billboard music charts and Variety’s streaming originals rankings. With Snowflake’s scalable data lake architecture, secure data sharing, Snowpark and Snowpark ML, Luminate increased daily data processing by 334%, runs market reports overnight instead of over a full month, and can build richer cross-industry insights for labels, studios, networks and partners.\u003C/p\u003E\r\n","richText":true,":type":"snowflake-site/components/text"},"layoutStyle":"horizontal","title":{"id":"title","type":"heading4","lines":["Customer story: Luminate"],":type":"snowflake-site/components/title-v2"},"button":{"id":"button","showOutboundIcon":false,"buttonLink":{"valid":true,"url":"/en/customers/all-customers/case-study/luminate/"},"linkTargetContentType":"DOCUMENT",":type":"snowflake-site/components/button","linkType":"SNOWFLAKE_INTERNAL","text":"Read the full case study"},"image":{"id":"image","height":"234","src":"https://www.snowflake.com/adobe/dynamicmedia/deliver/dm-aid--91a067a6-5d47-4e6d-9fdc-9bace95c71f0/luminate%25403x.png?quality=85&preferwebp=true","alt":"Luminate logo","lazyEnabled":true,"width":"416",":type":"snowflake-site/components/image"},"type":"content-card"},"text_ai-use-cases-built-on-behavioral-and-event-data_1":{"id":"text-fb433b1816","text":"\u003Ch3\u003ESports analytics AI use cases\u003C/h3\u003E\r\n\u003Ch4\u003EPlayer performance and workload management\u003C/h4\u003E\r\n\u003Cp\u003EUsed alongside professional medical guidance and coaching judgment, machine learning models can support decisions about training intensity, rotation and return to play. The data needs to remain connected to the correct athlete, session and competitive context. Performance measures can vary substantially by position, opponent and game state, so comparisons require more than a single aggregate score.\u003C/p\u003E\r\n\u003Ch4\u003EGame strategy and opponent analysis\u003C/h4\u003E\r\n\u003Cp\u003EAI can analyze play-by-play records, tracking data and video to identify tactical patterns, matchup tendencies and changes in opponent behavior. Coaches and analysts can use those findings to prepare game plans, evaluate lineups or understand how a strategy performs under particular conditions.\u003C/p\u003E\r\n\u003Ch4\u003EScouting and talent evaluation\u003C/h4\u003E\r\n\u003Cp\u003ETeams can use AI to compare player performance across leagues, competition levels and playing styles. Computer vision and event data can help surface movement, decision-making and role-specific traits that may not appear in traditional statistics. Reliable evaluation requires normalization across environments. Competition quality, team role, playing time and data availability can otherwise make players appear more comparable than they are.\u003C/p\u003E\r\n\u003Ch4\u003EFan engagement and personalization\u003C/h4\u003E\r\n\u003Cp\u003EA customer 360 can connect ticketing, digital engagement, merchandise, venue activity and content consumption to support more relevant fan experiences. Recommendation systems may personalize content, offers or communications according to the fan’s relationship with the team or event. Consent and channel preference should shape how this information is used.\u003C/p\u003E\r\n\u003Ch4\u003ETicketing and attendance forecasting\u003C/h4\u003E\r\n\u003Cp\u003EForecasting models can estimate demand by event, section and sales channel using schedule, opponent, timing, pricing and historical attendance. This information can support inventory allocation, staffing and promotional planning. External events, weather and schedule changes may also affect demand quickly.\u003C/p\u003E\r\n\u003Ch4\u003EVenue and event operations\u003C/h4\u003E\r\n\u003Cp\u003EAI can forecast entry volume, concession demand, staffing requirements and crowd movement across a venue. Computer vision and sensor data may also help identify congestion or operational issues during an event. Because venue layouts and event profiles differ, models need site-specific context and clearly defined escalation procedures. Wait times, staffing variance, inventory waste and incident response provide practical measures.\u003C/p\u003E\r\n\u003Ch4\u003EContent production and media operations\u003C/h4\u003E\r\n\u003Cp\u003EGenerative and multimodal AI can help tag highlights, search video archives, prepare clips and create localized or personalized content. Rights, sponsorship obligations and distribution windows need to remain attached to each asset so faster production doesn’t create unauthorized use.\u003C/p\u003E\r\n\u003Ch3\u003ECybersecurity AI use cases\u003C/h3\u003E\r\n\u003Cp\u003EAI-assisted security tools can support a range of use cases, from detecting threats and analyzing vulnerabilities to prioritizing alerts and assisting with incident response. Because these tools can produce false positives and false negatives, they should operate within clearly defined permissions, validation procedures and human-review requirements.\u003C/p\u003E\r\n\u003Ch4\u003EThreat and anomaly detection\u003C/h4\u003E\r\n\u003Cp\u003EMachine learning can identify unusual activity across identity, endpoint, network, application and cloud events, including combinations that don’t match a known signature. Because normal behavior changes, models need ongoing recalibration and should be measured through detection coverage, false positives and time to triage.\u003C/p\u003E\r\n\u003Ch4\u003EIdentity and access risk\u003C/h4\u003E\r\n\u003Cp\u003EAI can identify improbable travel, abnormal privilege use, unusual service-account activity and access patterns inconsistent with a user’s role, while graph analysis exposes paths to sensitive systems. Employment status, entitlements, device and authentication history help distinguish legitimate but uncommon activity from misuse.\u003C/p\u003E\r\n\u003Ch4\u003EAlert triage and investigation\u003C/h4\u003E\r\n\u003Cp\u003ESecurity agents can gather evidence from logs, threat intelligence and prior cases, summarize the event sequence and recommend investigative steps. Tool permissions should reflect the consequence of the action, with different approval thresholds for retrieving evidence, deactivating an account or isolating a host.\u003C/p\u003E\r\n\u003Ch4\u003EPhishing and social engineering detection\u003C/h4\u003E\r\n\u003Cp\u003ELanguage and multimodal models can evaluate email content, sender behavior, links, attachments and impersonation signals, and identify campaigns that vary wording while sharing infrastructure. Feedback from confirmed incidents improves detection as attackers adapt.\u003C/p\u003E\r\n\u003Ch4\u003EVulnerability prioritization\u003C/h4\u003E\r\n\u003Cp\u003EAI can combine severity, exploit activity, asset exposure, business criticality and existing controls to rank remediation work. Accurate asset inventory and ownership are prerequisites, since teams cannot remediate a finding without knowing whether the component is deployed and who owns it.\u003C/p\u003E\r\n\u003Ch4\u003EData loss and insider-risk detection\u003C/h4\u003E\r\n\u003Cp\u003EBehavioral models can identify unusual downloads, sharing activity, repository access or movement of sensitive information, while classification and lineage show what type of data is involved. Anomaly alone shouldn’t be treated as evidence of malicious intent; privacy controls, investigation procedures and business context remain necessary.\u003C/p\u003E\r\n\u003Ch4\u003EMalware and threat intelligence analysis\u003C/h4\u003E\r\n\u003Cp\u003EAI can classify malware, extract indicators, summarize technical reports and connect campaigns with infrastructure or tactics. Because threat intelligence may be incomplete, contradictory or deceptive, outputs should preserve source attribution, confidence and time sensitivity.\u003C/p\u003E\r\n\u003Ch4\u003ESecurity posture and control validation\u003C/h4\u003E\r\n\u003Cp\u003EAgents can gather configuration evidence, compare it with policy, and identify control gaps across cloud and software-as-a-service environments, then prepare remediation guidance or a reviewable change request. Findings are most useful when they map to a specific owner, configuration and policy exception with an expiry.\u003C/p\u003E\r\n\u003Cp\u003E\u003Ca href=\"https://www.snowflake.com/en/artificial-intelligence/industries/ai-in-cybersecurity/\"\u003ELearn more about AI in cybersecurity >\u003C/a\u003E\u003C/p\u003E\r\n","richText":true,":type":"snowflake-site/components/text","appliedCssClassNames":"text-color-text-05"},"title_why-industry-ai-runs-on-snowflake":{"id":"title-v2-60644b9717","additionalClasses":"anchor-title anchor-title--why-industry-ai-runs-on-snowflake","type":"heading2","lines":["Why industry AI runs on Snowflake"],":type":"snowflake-site/components/title-v2"},"text_why-industry-ai-runs-on-snowflake_0":{"id":"text-5a4eefe758","text":"\u003Cp\u003EThe applications in this directory use different data, models and operating processes, but they share several foundational implementation needs: governed access to current enterprise data, enough context to interpret it correctly and a controlled path from analysis to action.\u003C/p\u003E\n\u003Cp\u003ESnowflake brings structured, semi-structured and unstructured data together with AI capabilities in an environment that supports security and governance controls. \u003Ca href=\"https://www.snowflake.com/en/product/features/cortex/\"\u003ECortex AI\u003C/a\u003E includes capabilities for building agents that can use enterprise data, analyzing supported document and multimodal content, and providing natural-language access to governed information. Snowflake ML provides capabilities that support stages of predictive machine-learning workflows, including feature engineering, model development, deployment and monitoring.\u003C/p\u003E\n\u003Cp\u003EThrough \u003Ca href=\"https://www.snowflake.com/en/product/features/horizon/\"\u003EHorizon Catalog\u003C/a\u003E and related Snowflake capabilities, organizations can apply controls such as role-based access, masking policies, lineage and data-quality monitoring within supported workflows. Semantic models and contextual metadata can help AI systems use business definitions and relationships more consistently, rather than relying only on isolated tables or documents.\u003C/p\u003E\n\u003Cp\u003ESecure Data Sharing and Snowflake Data Clean Rooms can support governed collaboration while reducing the need to exchange separate copies of sensitive data.\u003C/p\u003E\n\u003Cp\u003EWith governed data, semantic context and AI infrastructure in place, teams may be able to reuse common capabilities across industry-specific applications while adapting them to each use case.\u003C/p\u003E","richText":true,":type":"snowflake-site/components/text","appliedCssClassNames":"text-color-text-05"}},":itemsOrder":["text_intro_0","title_what-shapes-ai-outcomes-across-industries","text_what-shapes-ai-outcomes-across-industries_0","title_ai-use-cases-built-on-sensitive-regulated-data","text_ai-use-cases-built-on-sensitive-regulated-data_0","card_v2_ai-use-cases-built-on-sensitive-regulated-data_0","text_ai-use-cases-built-on-sensitive-regulated-data_1","title_ai-use-cases-built-on-machine-and-operational-data","text_ai-use-cases-built-on-machine-and-operational-data_0","card_v2_ai-use-cases-built-on-machine-and-operational-data_0","title_ai-use-cases-built-on-documents-and-language","text_ai-use-cases-built-on-documents-and-language_0","title_ai-use-cases-built-on-behavioral-and-event-data","text_ai-use-cases-built-on-behavioral-and-event-data_0","card_v2_ai-use-cases-built-on-behavioral-and-event-data_0","text_ai-use-cases-built-on-behavioral-and-event-data_1","title_why-industry-ai-runs-on-snowflake","text_why-industry-ai-runs-on-snowflake_0"],"appliedCssClassNames":"snowflake-responsive-container-inner-padding-medium"},"flexible_column_content_container_2":{"additionalClasses":"hub-sidebar","layout":"SIMPLE","id":"hub-body-aside",":type":"snowflake-site/components/flexible-column-container/flexible-column-content-container",":items":{"container":{"additionalClasses":"sticky-sidebar","layout":"RESPONSIVE_GRID","columnCount":12,"columnClassNames":{"text_943981956_copy_":"aem-GridColumn aem-GridColumn--default--12","text_copy":"aem-GridColumn aem-GridColumn--default--12"},"gridClassNames":"aem-Grid aem-Grid--12 aem-Grid--default--12","id":"container-a85bb363b5",":type":"snowflake-site/components/container",":items":{"text_943981956_copy_":{"id":"text-a10e13ec52","additionalClasses":"eyebrow-text","text":"\u003Cp\u003EIn This Guide\u003C/p\u003E\r\n","richText":true,":type":"snowflake-site/components/text","appliedCssClassNames":"text-size-regular"},"text_copy":{"id":"text-3ff840dd12","additionalClasses":"page-toc","text":"\u003Cul\u003E\r\n\u003Cli data-anchor=\"what-shapes-ai-outcomes-across-industries\"\u003EWhat shapes AI outcomes across industries\u003C/li\u003E\r\n\u003Cli data-anchor=\"ai-use-cases-built-on-sensitive-regulated-data\"\u003EAI use cases built on sensitive, regulated data\u003C/li\u003E\r\n\u003Cli data-anchor=\"ai-use-cases-built-on-machine-and-operational-data\"\u003EAI use cases built on machine and operational data\u003C/li\u003E\r\n\u003Cli data-anchor=\"ai-use-cases-built-on-documents-and-language\"\u003EAI use cases built on documents and language\u003C/li\u003E\r\n\u003Cli data-anchor=\"ai-use-cases-built-on-behavioral-and-event-data\"\u003EAI use cases built on behavioral and event data\u003C/li\u003E\r\n\u003Cli data-anchor=\"why-industry-ai-runs-on-snowflake\"\u003EWhy industry AI runs on Snowflake\u003C/li\u003E\r\n\u003C/ul\u003E\r\n","richText":true,":type":"snowflake-site/components/text","appliedCssClassNames":"text-size-small text-color-text-05"}},":itemsOrder":["text_943981956_copy_","text_copy"],"appliedCssClassNames":"snowflake-responsive-container-inner-padding-medium"}},":itemsOrder":["container"],"appliedCssClassNames":"snowflake-responsive-container-inner-padding-small"},":type":"snowflake-site/components/flexible-column-container","isActiveTOC":false,"isBlogPage":false},"flexible_column_cont_1467213961":{"id":"flexible-column-container-fd29c64a29","propertiesId":"hub-explore-resources-header","type":"1-column","alignColumns":"top","containerMaxWidth":"extra-large","topPadding":"large","bottomPadding":"none","spaceBetween":"small","reverseOnMobile":false,"carouselOnMobile":false,"backgroundImageOption":"none","flexible_column_content_container_1":{"layout":"SIMPLE","id":"container-bf207c3bbf",":type":"snowflake-site/components/flexible-column-container/flexible-column-content-container",":items":{"title_v2":{"id":"title-v2-d8cd332728","additionalClasses":"hub-explore-resources-header__headline","type":"heading2","lines":["Explore AI Resources"],":type":"snowflake-site/components/title-v2","appliedCssClassNames":"left-alignment"}},":itemsOrder":["title_v2"]},":type":"snowflake-site/components/flexible-column-container","isActiveTOC":false,"isBlogPage":false,"appliedCssClassNames":"snowflake-flexible-column-container-gray-10-bg"},"flexible_column_cont_912630531":{"id":"flexible-column-container-63e2cd2532","propertiesId":"hub-explore-resources-grid","type":"1-column","alignColumns":"top","containerMaxWidth":"extra-large","topPadding":"small","bottomPadding":"large","spaceBetween":"small","reverseOnMobile":false,"carouselOnMobile":false,"backgroundImageOption":"none","flexible_column_content_container_1":{"layout":"SIMPLE","id":"hub-explore-resources-grid-inner",":type":"snowflake-site/components/flexible-column-container/flexible-column-content-container",":items":{"resource_chip_0":{"id":"content-chip-4800eba89b","tagText":"REPORT","tagColor":"#C6EDF1","cta":{"id":"cta","showOutboundIcon":false,"buttonLink":{"valid":true,"url":"https://www.snowflake.com/en/lp/radical-roi-generative-ai/"},"linkTargetContentType":"GENERIC",":type":"snowflake-site/components/button","linkType":"SNOWFLAKE_EXTERNAL","text":"Read more"},"headline":{"id":"title-v2-d31f58326d","type":"heading5","lines":["The ROI of Gen AI and Agents 2026"],":type":"snowflake-site/components/title-v2"},":type":"snowflake-site/components/content-chip","appliedCssClassNames":"snowflake-content-chip-white-bg"},"resource_chip_1":{"id":"content-chip-5982b41d40","tagText":"BLOG","tagColor":"#29B5E8","cta":{"id":"cta","showOutboundIcon":false,"buttonLink":{"valid":true,"url":"https://www.snowflake.com/en/blog/enterprise-ai-agents-agentic-ai/"},"linkTargetContentType":"GENERIC",":type":"snowflake-site/components/button","linkType":"SNOWFLAKE_EXTERNAL","text":"Read more"},"headline":{"id":"title","type":"heading5","lines":["Technological Revolution: Preparing for the Era of Agentic AI"],":type":"snowflake-site/components/title-v2"},":type":"snowflake-site/components/content-chip","appliedCssClassNames":"snowflake-content-chip-white-bg"},"resource_chip_2":{"id":"content-chip-81ca562c63","tagText":"BLOG","tagColor":"#29B5E8","cta":{"id":"cta","showOutboundIcon":false,"buttonLink":{"valid":true,"url":"https://www.snowflake.com/en/blog/ai-agents-healthcare-efficiency/"},"linkTargetContentType":"GENERIC",":type":"snowflake-site/components/button","linkType":"SNOWFLAKE_EXTERNAL","text":"Read more"},"headline":{"id":"title","type":"heading5","lines":["Modern Healthcare Organizations Use AI Agents to Unlock Workflow Efficiencies and Deliver Care-Critical Insights"],":type":"snowflake-site/components/title-v2"},":type":"snowflake-site/components/content-chip","appliedCssClassNames":"snowflake-content-chip-white-bg"},"resource_chip_3":{"id":"content-chip-b492ee5e39","tagText":"REPORT","tagColor":"#C6EDF1","cta":{"id":"cta","showOutboundIcon":false,"buttonLink":{"valid":true,"url":"https://www.snowflake.com/en/future-ai-interoperability-healthcare/"},"linkTargetContentType":"GENERIC",":type":"snowflake-site/components/button","linkType":"SNOWFLAKE_EXTERNAL","text":"Read more"},"headline":{"id":"title-v2-414add5de4","type":"heading5","lines":["The Future of AI + Interoperability in Healthcare"],":type":"snowflake-site/components/title-v2"},":type":"snowflake-site/components/content-chip","appliedCssClassNames":"snowflake-content-chip-white-bg"}},":itemsOrder":["resource_chip_0","resource_chip_1","resource_chip_2","resource_chip_3"]},":type":"snowflake-site/components/flexible-column-container","isActiveTOC":false,"isBlogPage":false,"appliedCssClassNames":"snowflake-flexible-column-container-gray-10-bg"},"flexible_column_cont_1158003461":{"id":"flexible-column-container-fb29bc6d25","propertiesId":"hub-explore-topics-header","type":"1-column","alignColumns":"top","containerMaxWidth":"extra-large","topPadding":"large","bottomPadding":"none","spaceBetween":"small","reverseOnMobile":false,"carouselOnMobile":false,"backgroundImageOption":"none","flexible_column_content_container_1":{"layout":"SIMPLE","id":"container-39df69aecb",":type":"snowflake-site/components/flexible-column-container/flexible-column-content-container",":items":{"title_v2_copy_copy":{"id":"title-v2-276eb29fd6","additionalClasses":"hub-explore-topics-header__headline","type":"heading2","lines":["Explore AI Topics"],":type":"snowflake-site/components/title-v2","appliedCssClassNames":"left-alignment"},"text_copy_copy":{"id":"text-634fb1764a","additionalClasses":"hub-explore-topics-header__subheadline","text":"\u003Cp\u003EDeep dives into every aspect of artificial intelligence\u003C/p\u003E\r\n","richText":true,":type":"snowflake-site/components/text","appliedCssClassNames":"text-color-text-05"}},":itemsOrder":["title_v2_copy_copy","text_copy_copy"],"appliedCssClassNames":"snowflake-responsive-container-inner-padding-extra-small"},":type":"snowflake-site/components/flexible-column-container","isActiveTOC":false,"isBlogPage":false,"appliedCssClassNames":"snowflake-flexible-column-container-white-bg"},"flexible_column_cont_1377146023":{"id":"flexible-column-container-bd712105ba","propertiesId":"hub-explore-topics-grid","type":"3-column-even","alignColumns":"top","containerMaxWidth":"extra-large","topPadding":"small","bottomPadding":"large","spaceBetween":"small","reverseOnMobile":false,"carouselOnMobile":false,"backgroundImageOption":"none","flexible_column_content_container_1":{"layout":"SIMPLE","id":"container-999f3035d8",":type":"snowflake-site/components/flexible-column-container/flexible-column-content-container",":items":{"topic_card_0":{"id":"card-v2-1c44b166d6","configurationStatus":{"configured":true,"message":""},":type":"snowflake-site/components/card-v2","text":{"id":"text","text":"\u003Cp\u003ELearn about the use of AI to analyze data, interpret information, generate content and take actions across business workflows.\u003C/p\u003E\r\n","richText":true,":type":"snowflake-site/components/text"},"layoutStyle":"vertical","title":{"id":"title","type":"heading4","lines":["AI in Business"],":type":"snowflake-site/components/title-v2"},"button":{"id":"button","showOutboundIcon":false,"buttonLink":{"valid":true,"url":"https://www.snowflake.com/en/artificial-intelligence/business/"},"linkTargetContentType":"GENERIC",":type":"snowflake-site/components/button","linkType":"SNOWFLAKE_EXTERNAL","text":"Learn more"},"type":"content-card"}},":itemsOrder":["topic_card_0"]},"flexible_column_content_container_2":{"layout":"SIMPLE","id":"container-7e94205462",":type":"snowflake-site/components/flexible-column-container/flexible-column-content-container",":items":{"topic_card_1":{"id":"card-v2-c81207d277","configurationStatus":{"configured":true,"message":""},":type":"snowflake-site/components/card-v2","text":{"id":"text","text":"\u003Cp\u003ELearn about the use of ML, NLP, behavioral analytics and related techniques to detect suspicious activity, prioritize risk and automate parts of security operations.\u003C/p\u003E\r\n","richText":true,":type":"snowflake-site/components/text"},"layoutStyle":"vertical","title":{"id":"title","type":"heading4","lines":["AI in Cybersecurity"],":type":"snowflake-site/components/title-v2"},"button":{"id":"button","showOutboundIcon":false,"linkTargetContentType":"GENERIC",":type":"snowflake-site/components/button","linkType":"SNOWFLAKE_INTERNAL","text":"Learn more"},"type":"content-card"}},":itemsOrder":["topic_card_1"]},"flexible_column_content_container_3":{"layout":"SIMPLE","id":"container-b49f33335c",":type":"snowflake-site/components/flexible-column-container/flexible-column-content-container",":items":{"topic_card_2":{"id":"card-v2-300b597e9e","configurationStatus":{"configured":true,"message":""},":type":"snowflake-site/components/card-v2","text":{"id":"text","text":"\u003Cp\u003ELearn about AI-driven personalization, demand forecasting and supply chain optimization for retail and ecommerce.\u003C/p\u003E\r\n","richText":true,":type":"snowflake-site/components/text"},"layoutStyle":"vertical","title":{"id":"title","type":"heading4","lines":["AI in Retail"],":type":"snowflake-site/components/title-v2"},"button":{"id":"button","showOutboundIcon":false,"buttonLink":{"valid":true,"url":"https://www.snowflake.com/en/artificial-intelligence/industries/ai-in-retail/"},"linkTargetContentType":"GENERIC",":type":"snowflake-site/components/button","linkType":"SNOWFLAKE_EXTERNAL","text":"Learn more"},"type":"content-card"}},":itemsOrder":["topic_card_2"]},":type":"snowflake-site/components/flexible-column-container","isActiveTOC":false,"isBlogPage":false,"appliedCssClassNames":"snowflake-flexible-column-container-white-bg"}},":itemsOrder":["flexible_column_cont","flexible_column_cont_939100716","flexible_column_cont_1398138236","flexible_column_cont_663228916","flexible_column_cont_1467213961","flexible_column_cont_912630531","flexible_column_cont_1158003461","flexible_column_cont_1377146023"],":type":"wcm/foundation/components/responsivegrid"},"modal_container":{"layout":"SIMPLE","id":"container-9c2d80cada",":type":"snowflake-site/components/modal/modal-container",":items":{},":itemsOrder":[]},"markup_editor_928258845":{"id":"markup-editor-6f2ec806fe","title":" ","cssContent":".snowflake-flexible-column-container-gray-10-bg\u003E.snowflake-flexible-column-container{background-color:var(--ui-background-05) !important}.text-size-regular:has(.seo-hub-hero__related-topic-label){display:flex;align-items:center}.hub-hero__headline span{text-transform:none !important}.hub-hero__subheadline p{max-width:70ch;margin-top:8px}.hub-hero__authors \u003E .container \u003E .cmp-container \u003E .aem-container{display:flex;flex-direction:row}.hub-hero__authors \u003E .container \u003E .cmp-container \u003E .aem-container \u003E div{width:auto !important;margin:24px 48px 0 0 !important}.hub-hero__authors .heading-5-v2{gap:var(--spacing-00)}.hub-hero__authors .snowflake-content-chip-button{display:none !important}.hub-hero__authors .snowflake-person-chip-content .body-2,.hub-hero__authors .snowflake-content-chip-content .snowflake-title-v2-line{font-size:16px !important;line-height:20px !important;font-family:\"Lato\",sans-serif !important;color:#000 !important;font-weight:600 !important}.hub-hero__authors .snowflake-person-chip-content .body-3,.hub-hero__authors .snowflake-content-chip-content .snowflake-title-v2-line:not(:first-child){font-weight:400 !important;color:var(--text-05) !important;font-size:16px !important}.hub-hero__authors .snowflake-image-container img{aspect-ratio:1 !important;border-radius:100%;overflow:hidden}.hub-hero__authors .snowflake-person-chip-avatar{width:56px;height:56px}.hub-hero__authors .snowflake-content-chip{align-items:center;display:inline-flex}.hub-hero__authors .snowflake-content-chip-image{max-width:56px;line-height:0;margin-right:var(--spacing-03)}.hub-hero__authors .snowflake-person-chip-inner-horizontal{gap:var(--spacing-03)}@media screen and (min-width:1367px){.hub-hero__headline .heading-1-v2{font-size:48px;line-height:44px}}#hub-explore-resources-grid-inner\u003E.container\u003E.cmp-container\u003E.aem-container{display:flex;flex-direction:row;flex-wrap:wrap;gap:24px}#hub-explore-resources-grid-inner\u003E.container\u003E.cmp-container\u003E.aem-container::before,#hub-explore-resources-grid-inner\u003E.container\u003E.cmp-container\u003E.aem-container::after{display:none}#hub-explore-resources-grid-inner\u003E.container\u003E.cmp-container\u003E.aem-container\u003Ediv{width:calc(25% - 18px)}.text-color-text-05 .snowflake-text h2,.text-color-text-05.cq-Editable-dom h2,.text-color-text-05 .snowflake-text h3,.text-color-text-05.cq-Editable-dom h3,.text-color-text-05 .snowflake-text h4,.text-color-text-05.cq-Editable-dom h4,.text-color-text-05 .snowflake-text h5,.text-color-text-05.cq-Editable-dom h5,.text-color-text-05 .snowflake-text h6,.text-color-text-05.cq-Editable-dom h6{color:#000 !important}",":type":"snowflake-site/components/markup-editor","isGSAPEnabled":false},"markup_editor_597730182":{"id":"markup-editor-7ff9b4f8e7","title":" ","cssContent":".sf-copy-markdown [data-copy-md]{display:inline-flex;align-items:center;gap:6px;padding:8px 16px;font-family:'Texta',sans-serif;text-transform:uppercase;font-weight:800 !important;font-size:14px;font-weight:500;color:#11567f;background:#f0faff;border:1px solid #b8e6f9;border-radius:24px;cursor:pointer;transition:background .2s ease,border-color .2s ease,color .2s ease}.sf-copy-markdown [data-copy-md]:hover{background:#ddf3fc;border-color:#29b5e8}.sf-copy-markdown [data-copy-md][data-copied=\"1\"]{color:#0f7b3e;background:#ecfdf5;border-color:#6ee7a0;pointer-events:none}.sf-copy-markdown [data-copy-md] svg{flex-shrink:0}.longform-conten .snowflake-content-chip-white-bg .snowflake-content-chip{box-shadow:0 0 24px 4px rgba(0,0,0,.02),0 4px 8px 0 rgba(0,0,0,.04);flex-direction:row-reverse;align-items:center}.longform-conten .snowflake-content-chip-button{display:none}.longform-conten .snowflake-content-chip-image__inner{aspect-ratio:5 / 3;display:flex;justify-content:center;align-items:center;background-color:var(--ui-01);border-radius:4px}.longform-conten .snowflake-content-chip-image{margin-right:0;margin-left:48px}.longform-conten .snowflake-content-chip-image img{width:50%;border-radius:0 !important;object-fit:contain}.longform-content .black-blue-text-color .snowflake-title-v2-line:not(:first-child){font-size:14px !important;font-weight:400 !important;color:rgba(0,0,0,.6) !important;margin-top:8px !important}.page-toc ul li:first-child{padding-top:0 !important}.page-toc ul li:last-child{padding-bottom:0 !important}.seo-hub__top-bar \u003E .container \u003E .cmp-container \u003E .aem-container \u003E div:first-child{flex-grow:1}.sf-copy-markdown{margin-top:40px !important}.page-toc ul{margin-top:16px !important}.seo-hub__top-bar \u003E .container \u003E .cmp-container \u003E .aem-container{display:flex;justify-content:space-between}.sf-copy-markdown{}.callout.snowflake-text p:not(:first-child){margin-top:var(--spacing-01)}.callout \u003E span \u003E p:first-child \u003E strong,.callout \u003E span \u003E p:first-child \u003E b{text-transform:uppercase;font-family:'Texta',sans-serif;font-size:16px !important;color:var(--ui-01) !important}.seo-hub-hero__subheadline p{max-width:50ch}.tag-group ul{list-style-type:none;padding:0;margin:0;display:flex;flex-direction:row;row-gap:12px;column-gap:8px;align-items:center;flex-wrap:wrap}.tag-group ul li:first-child{flex-shrink:0}.tag-group ul li a{display:inline-block;padding:2px 12px;border-radius:48px;background-color:#ededed;color:#666;font-size:14px !important}@media screen and (min-width:1367px){.seo-hub-hero__headline span.snowflake-title-v2-line{font-size:56px !important}}.callout.snowflake-text p:not(:first-child){margin-top:var(--spacing-01)}.callout \u003E span \u003E p:first-child \u003E b{text-transform:uppercase;font-family:'Texta',sans-serif;font-size:16px !important;color:var(--ui-01) !important}.seo-hub-hero__subheadline p{max-width:80ch}#hero:has(.snowflake-youtube-lite) .seo-hub-hero__subheadline p{max-width:50ch}.tag-group ul{list-style-type:none;padding:0;margin:0;display:flex;flex-direction:row;row-gap:12px;column-gap:8px;align-items:center;flex-wrap:wrap}.tag-group ul li:first-child{width:100%;flex-shrink:0}.tag-group ul li a{display:inline-block;padding:2px 12px;border-radius:48px;background-color:#ededed;color:#666;font-size:14px !important}@media screen and (min-width:1367px){.seo-hub-hero__headline span.snowflake-title-v2-line{font-size:56px !important}}",":type":"snowflake-site/components/markup-editor","isGSAPEnabled":false},"markup_editor":{"id":"markup-editor-9a8f14dd75","title":" ","cssContent":"div.snowflake-breadcrumb a.snowflake-breadcrumb-item,.snowflake-breadcrumb div.snowflake-breadcrumb-item{text-transform:none;font-weight:500}.snowflake-breadcrumb svg{display:none !important}.snowflake-breadcrumb a:has(svg)::after{content:'/';margin:0 12px;color:#666}.hub-sidebar{padding:0 40px}.sticky-sidebar{max-width:340px;margin-left:auto}.page-toc ul{list-style-type:none;padding:0}.page-toc li{padding:8px 16px;border-left:4px solid var(--ui-01);cursor:pointer;transition:300ms ease all}.page-toc li:hover{color:var(--ui-01);border-color:#7fd3f1;transition:300ms ease all}.callout,.customer-card{background-color:#eef9fd;border-left:4px solid var(--ui-01);padding:24px 24px 24px 32px;border-radius:4px}.logo-container{max-width:180px}.longform-content li{margin-top:1rem !important}div.longform-content p{max-width:80ch}.bolder .snowflake-title-v2-line{font-weight:900 !important}.border-top\u003Ediv{border-top:1px solid #ccc;padding-top:48px}.related-topics ul{list-style-type:none;padding:0;margin:0;display:flex;gap:8px;flex-wrap:wrap}.related-topics li{display:inline-block;border:1px solid #ccc;padding:4px 12px;border-radius:24px}div.longform-content .snowflake-text h2,div.longform-content .snowflake-text .heading-2-v2,div.longform-content .snowflake-text h3,div.longform-content .snowflake-text .heading-3-v2,div.longform-content .snowflake-title-v2 .heading-3-v2,div.longform-content .snowflake-text h4,div.longform-content .snowflake-text .heading-4-v2,div.longform-content .snowflake-title-v2 .heading-4-v2,div.longform-content .snowflake-text h5,div.longform-content .snowflake-text .heading-5-v2,div.longform-content .snowflake-title-v2 .heading-5-v2,div.longform-content .snowflake-text h6,div.longform-content .snowflake-title-v2 .heading-6-v2,div.longform-content .snowflake-text .heading-6-v2{text-transform:none !important}div.longform-content .snowflake-text h2,div.longform-content .snowflake-text .heading-2-v2,div.longform-content .snowflake-text h3,div.longform-content .snowflake-text .heading-3-v2,div.longform-content .snowflake-text h4,div.longform-content .snowflake-text .heading-4-v2,div.longform-content .snowflake-text h5,div.longform-content .snowflake-text .heading-5-v2,div.longform-content .snowflake-text h6,div.longform-content .snowflake-text .heading-6-v2{margin-top:1.5rem !important;line-height:1.1 !important}div.longform-content .snowflake-text h3,div.longform-content .snowflake-text .heading-3-v2,div.longform-content .snowflake-title-v2 .heading-3-v2,div.longform-content .snowflake-text h4,div.longform-content .snowflake-text .heading-4-v2,div.longform-content .snowflake-title-v2 .heading-4-v2,div.longform-content .snowflake-text h5,div.longform-content .snowflake-text .heading-5-v2,div.longform-content .snowflake-title-v2 .heading-5-v2,div.longform-content .snowflake-text h6,div.longform-content .snowflake-text .heading-6-v2,div.longform-content .snowflake-title-v2 .heading-6-v2{font-family:Lato,sans-serif !important;font-weight:800 !important}div.longform-content .snowflake-text h2,div.longform-content .snowflake-text .heading-2-v2,div.longform-content .snowflake-title-v2 .heading-2-v2{text-transform:none !important;font-size:28px !important}div.longform-content .snowflake-text h3,div.longform-content .snowflake-text .heading-3-v2,div.longform-content .snowflake-title-v2 .heading-3-v2{font-size:22px !important}div.longform-content .snowflake-text h4,div.longform-content .snowflake-text .heading-4-v2,div.longform-content .snowflake-title-v2 .heading-4-v2{font-size:18px !important}div.longform-content .snowflake-text h5,div.longform-content .snowflake-text .heading-5-v2,div.longform-content .snowflake-title-v2 .heading-5-v2{font-size:16px !important}div.longform-content .snowflake-text h6,div.longform-content .snowflake-text .heading-6-v2,div.longform-content .snowflake-title-v2 .heading-6-v2{font-size:14px !important}@media screen and (min-width:992px){div.longform-content .snowflake-text h2,div.longform-content .snowflake-text .heading-2-v2,div.longform-content .snowflake-title-v2 .heading-2-v2{font-size:38px !important}div.longform-content .snowflake-text h3,div.longform-content .snowflake-text .heading-3-v2,div.longform-content .snowflake-title-v2 .heading-3-v2{font-size:26px !important}div.longform-content .snowflake-text h4,div.longform-content .snowflake-text .heading-4-v2,div.longform-content .snowflake-title-v2 .heading-4-v2{font-size:22px !important}div.longform-content .snowflake-text h5,div.longform-content .snowflake-text .heading-5-v2,div.longform-content .snowflake-title-v2 .heading-5-v2{font-size:18px !important}div.longform-content .snowflake-text h6,div.longform-content .snowflake-text .heading-6-v2,div.longform-content .snowflake-title-v2 .heading-6-v2{font-size:16px !important}}.sticky-sidebar .page-toc li.is-active{font-weight:600;color:var(--snow-blue,#29b5e8)}.sticky-sidebar .page-toc li[data-anchor]{cursor:pointer}.longform-content table{margin-top:24px;margin-bottom:24px;width:100%;background-color:var(--ui-background-01);border-collapse:collapse;border:2px solid var(--ui-background-09);font-family:'Lato',sans-serif;color:var(--ui-background-09)}.longform-content table thead{background-color:var(--ui-01)}.longform-content th,.longform-content td{min-width:120px;border:2px solid var(--ui-background-09);padding:var(--spacing-01)}.longform-content ol{margin-top:0 !important}.longform-content ol li{margin-bottom:1rem !important}.longform-content ul li{margin:0;padding:0 0 0 32px;position:relative}.longform-content ul{list-style-type:none}.longform-content ul li::before{content:\"\";display:block;border-radius:100%;background:#29b5e8;width:18px;height:18px;position:absolute;top:4px;left:0;border:5px solid #e5f2f7;box-sizing:border-box}.seo-customer.snowflake-card-v2-advanced-horizontal .snowflake-card-v2-advanced-image-container{max-width:200px}.seo-customer.snowflake-card-v2-advanced-horizontal .snowflake-card-v2-advanced-image-container img{object-fit:contain}.related-topics-outer-container \u003E .container \u003E .cmp-container \u003E .aem-container{display:flex;flex-direction:row}.related-topics-outer-container \u003E .container \u003E .cmp-container \u003E .aem-container \u003E div{width:auto !important;margin:0 !important}.related-topics-outer-container \u003E .container \u003E .cmp-container \u003E .aem-container \u003E div:first-child{margin-right:16px !important;flex-shrink:0}","jsContent":"(function(){var OFFSET=100;if(window.gsap&&window.ScrollTrigger){gsap.registerPlugin(ScrollTrigger);var sidebar=document.querySelector('.sticky-sidebar');var body=document.querySelector('.longform-content');if(sidebar&&body){ScrollTrigger.create({trigger:sidebar,start:'top 100px',endTrigger:body,end:'bottom bottom',pin:sidebar,pinSpacing:false});}}document.addEventListener('click',function(e){var li=e.target.closest('li[data-anchor]');if(!li)return;var slug=li.getAttribute('data-anchor');var heading=document.querySelector('.anchor-title--'+CSS.escape(slug));if(!heading)return;e.preventDefault();var top=heading.getBoundingClientRect().top+window.pageYOffset-OFFSET;window.scrollTo({top:top,behavior:'smooth'});history.replaceState(null,'','#'+slug);},false);var headings=document.querySelectorAll('[class*=\"anchor-title--\"]');if(headings.length&&'IntersectionObserver'in window){var io=new IntersectionObserver(function(entries){entries.forEach(function(entry){if(!entry.isIntersecting)return;var cls=Array.from(entry.target.classList).find(function(c){return c.indexOf('anchor-title--')===0;});if(!cls)return;var slug=cls.replace('anchor-title--','');document.querySelectorAll('li[data-anchor]').forEach(function(li){li.classList.toggle('is-active',li.getAttribute('data-anchor')===slug);});});},{rootMargin:'-20% 0px -70% 0px',threshold:0});headings.forEach(function(h){io.observe(h);});}})();",":type":"snowflake-site/components/markup-editor","isGSAPEnabled":true},"experiencefragment-footer":{"id":"experiencefragment-ef20bc9751","localizedFragmentVariationPath":"/content/experience-fragments/snowflake-site/language-masters/en/site/footer/master/jcr:content","configured":true,":type":"snowflake-site/components/experiencefragment","xfModelPath":"/content/experience-fragments/snowflake-site/language-masters/en/site/footer/master.xfmodel.json"},"experiencefragment":{"id":"experiencefragment-1d39e8c422","localizedFragmentVariationPath":"/content/experience-fragments/snowflake-site/language-masters/en/site/footer-legal-disclaimers/master/jcr:content","configured":true,":type":"snowflake-site/components/experiencefragment","xfModelPath":"/content/experience-fragments/snowflake-site/language-masters/en/site/footer-legal-disclaimers/master.xfmodel.json"}},":itemsOrder":["experiencefragment-banner","experiencefragment-header","responsivegrid","modal_container","markup_editor_928258845","markup_editor_597730182","markup_editor","experiencefragment-footer","experiencefragment"],":type":"wcm/foundation/components/responsivegrid"}},":itemsOrder":["root"],"isPasswordProtected":false,"analyticsContentTags":["snowflake-site:taxonomy/content-type/fundamentals"],"analyticsEnabled":true,":hierarchyType":"page",":path":"/content/snowflake-site/global/en/artificial-intelligence/industries","coveoConfig":{"pipeline":"snowflake.com","apiKey":"xx335921a6-2a0a-40f2-a167-e390b4766c3d","organizationId":"snowflakecomputingproduction8neljofn","searchHub":"snowflake.com"},"analyticsDebugMode":false,"analyticsData":{"excludeFromAnalytics":false,"subCategory":"","pageType":"homepage","templateName":"base-page-template54","siteName":"snowflake","pageUrl":"/content/snowflake-site/global/en/artificial-intelligence/industries","language":"en","category":"general","pageName":"AI in Action: A Directory of AI Use Cases Across Industries","contentTags":["snowflake-site:taxonomy/content-type/fundamentals"]},"locale":"en"}