CINXE.COM
What are AI guardrails? | McKinsey
<!DOCTYPE html><html dir="ltr" lang="en"><head><meta charSet="utf-8"/><meta name="viewport" content="width=device-width, initial-scale=1.0"/><meta http-equiv="X-UA-Compatible" content="IE=edge"/><script>var McKinsey = {"ArticleTemplate":"Legacy","DaysSinceCMSPublication":"7","DisplayDate":"11/14/2024","OriginalPublicationDate":"11/14/2024","SitecoreId":"{3899126A-E938-4641-BB09-FC15C3868982}","Title":"What are AI guardrails?","ArticleType":"Article","ContentType":"Article","ServerNumber":"","IsPageRestricted":"true","UserID":"","RegistrationDate":"","LoginStatus":"logged_out","JobTitle":"","CompanyName":"","blogTags":null,"enableRegWall":false}; var pageMetaInformation = {"CurrentLanguage":"en","AlternateLanguages":[{"DisplayName":"English","LanguageCode":"en","Url":"/featured-insights/mckinsey-explainers/what-are-ai-guardrails"}],"NavigationLink":"featured-insights","ActiveItemId":"{3B4756D0-2A01-44CA-B37A-B4CA73D83717}","OfficeCode":"","MiniSiteId":""};</script><link rel="icon" href="/favicon.ico"/><link href="https://www.mckinsey.com/redesign/resources/css/styles-rc.css" rel="stylesheet"/><link rel="manifest" href="/manifest.json"/><link rel="dns-prefetch" href="//cdn.dynamicyield.com"/><link rel="dns-prefetch" href="//st.dynamicyield.com"/><link rel="dns-prefetch" href="//rcom.dynamicyield.com"/><link rel="dns-prefetch" href="//cdn.cookielaw.org"/><link rel="preconnect" href="//assets.adobedtm.com"/><link rel="preconnect" href="//connect.facebook.net"/><link rel="preconnect" href="//static.hotjar.com"/><link rel="preload" as="font" href="/next-static/fonts/bower/Bower-Bold.woff2" type="font/woff2" crossorigin=""/><link rel="preload" as="font" href="/next-static/fonts/mckinsey-sans/regular/McKinseySans-Regular.woff2" type="font/woff2" crossorigin=""/><link rel="preload" as="font" href="/next-static/fonts/mckinsey-sans/medium/McKinseySans-Medium.woff2" type="font/woff2" crossorigin=""/><link rel="preload" as="font" href="/next-static/fonts/mckinsey-sans/light/McKinseySans-Light.woff2" type="font/woff2" crossorigin=""/><link rel="preload" as="font" href="/next-static/fonts/mckinsey-sans/italic/McKinseySans-Italic.woff2" type="font/woff2" crossorigin=""/><link rel="preload" as="font" href="/next-static/fonts/mckinsey-sans/light-italic/McKinseySans-LightItalic.woff2" type="font/woff2" crossorigin=""/><link rel="preload" as="font" href="/next-static/fonts/mckinsey-sans/medium-italic/McKinseySans-MediumItalic.woff2" type="font/woff2" crossorigin=""/><link rel="preload" as="font" href="/next-static/fonts/noto-sans-jp/NotoSansJP-Light.ttf" type="font/ttf" crossorigin=""/><link rel="preload" as="font" href="/next-static/fonts/noto-sans-jp/NotoSansJP-Regular.ttf" type="font/ttf" crossorigin=""/><link rel="preload" as="font" href="/next-static/fonts/noto-sans-jp/NotoSansJP-Medium.ttf" type="font/ttf" crossorigin=""/><link rel="preload" as="font" href="/next-static/fonts/noto-sans-jp/NotoSansJP-Bold.ttf" type="font/ttf" crossorigin=""/><link rel="preload" as="font" href="/next-static/fonts/noto-sans-jp/NotoSansJP-Black.ttf" type="font/ttf" crossorigin=""/><link rel="preload" as="font" href="/next-static/fonts/noto-serif-jp/NotoSerifJP-Bold.ttf" type="font/ttf" crossorigin=""/><meta name="apple-itunes-app" content="app-id=674902075"/><link rel="apple-touch-icon" sizes="57x57" href="/next-static/images/mck-touch-icon-57x57.png"/><link rel="apple-touch-icon" sizes="72x72" href="/next-static/images/mck-touch-icon-72x72.png"/><link rel="apple-touch-icon" sizes="114x114" href="/next-static/images/mck-touch-icon-114x114.png"/><link rel="apple-touch-icon" sizes="144x144" href="/next-static/images/mck-touch-icon-144x144.png"/><link rel="apple-touch-icon" sizes="152x152" href="/next-static/images/mck-touch-icon-152x152.png"/><link rel="apple-touch-icon" sizes="167x167" href="/next-static/images/mck-touch-icon-167x167.png"/><link rel="apple-touch-icon" sizes="180x180" href="/next-static/images/mck-touch-icon-180x180.png"/><meta content="no-referrer-when-downgrade" name="referrer"/><meta content="McKinsey & Company" property="og:site_name" name="site_name"/><meta content="What are AI guardrails?" property="og:title" name="title"/><meta content="AI guardrails help ensure that an organization’s AI tools, and their application in the business, reflect the organization’s standards, policies, and values." property="og:description" name="description"/><meta content="https://www.mckinsey.com/featured-insights/mckinsey-explainers/what-are-ai-guardrails" name="url" property="og:url"/><meta content="index,follow,all" name="robots"/><meta content="https://www.mckinsey.com/~/media/mckinsey/featured%20insights/mckinsey%20explainers/what%20are%20ai%20guardrails/what%20are%20ai%20guardrails_-2152543981-thumb-1536x1536.jpg" property="og:image" name="image"/><meta content="{3899126A-E938-4641-BB09-FC15C3868982}" name="sid"/><link href="https://www.mckinsey.com/featured-insights/mckinsey-explainers/what-are-ai-guardrails" rel="canonical"/><meta content="Monthly Highlights" name="practice-name"/><meta content="N02" name="practice-code"/><meta content="summary_large_image" name="twitter:card"/><meta content="@mckinsey" name="twitter:site"/><title>What are AI guardrails? | McKinsey</title><meta content="Insights & Publications" name="sections"/><meta content="What are AI guardrails?" name="twitter:title"/><meta content="AI guardrails help ensure that an organization’s AI tools, and their application in the business, reflect the organization’s standards, policies, and values." name="twitter:description"/><meta content="Article" property="contenttype" name="contenttype"/><meta content="https://www.mckinsey.com/~/media/mckinsey/featured%20insights/mckinsey%20explainers/what%20are%20ai%20guardrails/what%20are%20ai%20guardrails_-2152543981-thumb-1536x1536.jpg?mw=677&car=42:25" name="twitter:image"/><meta content="What are AI guardrails?" name="twitter:image:alt"/><meta content="true" name="mobileready"/><meta content="public" name="accesslevel"/><meta content="false" name="excludefromclientlink"/><meta content="article" name="articletype"/><meta content="2024-11-14T00:00:00Z" name="itemdate"/><meta content="Featured Insights | Mckinsey Explainers | Article | November 14, 2024" name="searchresults-tags"/><script type="application/ld+json">{"@context":"https://schema.org","@type":"Article","url":"https://www.mckinsey.com","publisher":{"@type":"Organization","name":"McKinsey & Company","logo":{"@type":"ImageObject","url":"https://www.mckinsey.com/~/media/Thumbnails/Mck_Logo"}},"mainEntityOfPage":{"@type":"WebPage","@id":"https://www.mckinsey.com/featured-insights/mckinsey-explainers/what-are-ai-guardrails"},"datePublished":"2024-11-14T10:00:00Z","dateCreated":"2024-11-12T18:46:26Z","dateModified":"2024-11-14T00:00:00Z","heading":"What are AI guardrails?","image":"https://www.mckinsey.com/~/media/mckinsey/featured%20insights/mckinsey%20explainers/what%20are%20ai%20guardrails/what%20are%20ai%20guardrails_-2152543981-thumb-1536x1536.jpg","description":"AI guardrails help ensure that an organization’s AI tools, and their application in the business, reflect the organization’s standards, policies, and values."}</script><meta name="next-head-count" content="62"/><meta name="next-font-preconnect"/><script src="" id="onetrust-wrapperchecker" data-nscript="beforeInteractive">function OptanonWrapperChecker() {}</script><link rel="preload" href="/_next/static/css/e0c1ca92b5263a45.css" as="style"/><link rel="stylesheet" href="/_next/static/css/e0c1ca92b5263a45.css" data-n-g=""/><link rel="preload" href="/_next/static/css/985443d8095c0b48.css" as="style"/><link rel="stylesheet" href="/_next/static/css/985443d8095c0b48.css" data-n-p=""/><link rel="preload" href="/_next/static/css/b9ab7b649e8aa36e.css" as="style"/><link rel="stylesheet" href="/_next/static/css/b9ab7b649e8aa36e.css"/><link rel="preload" href="/_next/static/css/da15493cadea4a2c.css" as="style"/><link rel="stylesheet" href="/_next/static/css/da15493cadea4a2c.css"/><link rel="preload" href="/_next/static/css/72fc56b1fbe032aa.css" as="style"/><link rel="stylesheet" href="/_next/static/css/72fc56b1fbe032aa.css"/><noscript data-n-css=""></noscript><script defer="" nomodule="" src="/_next/static/chunks/polyfills-5cd94c89d3acac5f.js"></script><script src="https://cdn.cookielaw.org/scripttemplates/otSDKStub.js" data-document-language="true" type="text/javascript" data-domain-script="915b5091-0d7e-44d2-a8c4-cf08267e52fe" defer="" data-nscript="beforeInteractive"></script><script defer="" src="/_next/static/chunks/281.cc3ac8578a543e8e.js"></script><script defer="" src="/_next/static/chunks/5228.25594813f55e8bf1.js"></script><script defer="" src="/_next/static/chunks/390.5f3d9ea01929a8fb.js"></script><script defer="" src="/_next/static/chunks/6731.9bccca8a7edb0704.js"></script><script defer="" src="/_next/static/chunks/2354.05223bcb67691d75.js"></script><script defer="" src="/_next/static/chunks/2991.4cb0aa2f9f53e653.js"></script><script src="/_next/static/chunks/webpack.7b31c4548c626f3b.js" defer=""></script><script src="/_next/static/chunks/framework.62bbe2ca94854a85.js" defer=""></script><script src="/_next/static/chunks/main.51e10588adc949ca.js" defer=""></script><script src="/_next/static/chunks/pages/_app.5f99628c3d07543d.js" defer=""></script><script src="/_next/static/chunks/3b1baa31.cd6cdac6158774d8.js" defer=""></script><script src="/_next/static/chunks/7d0bf13e.8f3383787afb45af.js" defer=""></script><script src="/_next/static/chunks/1354.c34ddc4bd7c986c8.js" defer=""></script><script src="/_next/static/chunks/408.2780008a4c2ee197.js" defer=""></script><script src="/_next/static/chunks/pages/%5B%5B...path%5D%5D.eb738e7c87df0dc8.js" defer=""></script><script src="/_next/static/9vYT_fWQKRjfPT7lAZJdB/_buildManifest.js" defer=""></script><script src="/_next/static/9vYT_fWQKRjfPT7lAZJdB/_ssgManifest.js" defer=""></script><script src="/_next/static/9vYT_fWQKRjfPT7lAZJdB/_middlewareManifest.js" defer=""></script> <script>(window.BOOMR_mq=window.BOOMR_mq||[]).push(["addVar",{"rua.upush":"false","rua.cpush":"false","rua.upre":"false","rua.cpre":"false","rua.uprl":"false","rua.cprl":"false","rua.cprf":"false","rua.trans":"","rua.cook":"false","rua.ims":"false","rua.ufprl":"false","rua.cfprl":"false","rua.isuxp":"false","rua.texp":"norulematch","rua.ceh":"false","rua.ueh":"false","rua.ieh.st":"0"}]);</script> <script>!function(a){var e="https://s.go-mpulse.net/boomerang/",t="addEventListener";if("False"=="True")a.BOOMR_config=a.BOOMR_config||{},a.BOOMR_config.PageParams=a.BOOMR_config.PageParams||{},a.BOOMR_config.PageParams.pci=!0,e="https://s2.go-mpulse.net/boomerang/";if(window.BOOMR_API_key="TURRK-8ADJT-WDUC5-TC32E-KV9ND",function(){function n(e){a.BOOMR_onload=e&&e.timeStamp||(new Date).getTime()}if(!a.BOOMR||!a.BOOMR.version&&!a.BOOMR.snippetExecuted){a.BOOMR=a.BOOMR||{},a.BOOMR.snippetExecuted=!0;var i,_,o,r=document.createElement("iframe");if(a[t])a[t]("load",n,!1);else if(a.attachEvent)a.attachEvent("onload",n);r.src="javascript:void(0)",r.title="",r.role="presentation",(r.frameElement||r).style.cssText="width:0;height:0;border:0;display:none;",o=document.getElementsByTagName("script")[0],o.parentNode.insertBefore(r,o);try{_=r.contentWindow.document}catch(O){i=document.domain,r.src="javascript:var d=document.open();d.domain='"+i+"';void(0);",_=r.contentWindow.document}_.open()._l=function(){var a=this.createElement("script");if(i)this.domain=i;a.id="boomr-if-as",a.src=e+"TURRK-8ADJT-WDUC5-TC32E-KV9ND",BOOMR_lstart=(new Date).getTime(),this.body.appendChild(a)},_.write("<bo"+'dy onload="document._l();">'),_.close()}}(),"".length>0)if(a&&"performance"in a&&a.performance&&"function"==typeof a.performance.setResourceTimingBufferSize)a.performance.setResourceTimingBufferSize();!function(){if(BOOMR=a.BOOMR||{},BOOMR.plugins=BOOMR.plugins||{},!BOOMR.plugins.AK){var e=""=="true"?1:0,t="",n="bdpnbeqxgy4diz2g2a5q-f-82cdcb4cc-clientnsv4-s.akamaihd.net",i="false"=="true"?2:1,_={"ak.v":"39","ak.cp":"19387","ak.ai":parseInt("285213",10),"ak.ol":"0","ak.cr":4,"ak.ipv":4,"ak.proto":"http/1.1","ak.rid":"41b6cf2c","ak.r":37669,"ak.a2":e,"ak.m":"a","ak.n":"essl","ak.bpcip":"8.222.208.0","ak.cport":45192,"ak.gh":"23.53.33.206","ak.quicv":"","ak.tlsv":"tls1.2","ak.0rtt":"","ak.0rtt.ed":"","ak.csrc":"-","ak.acc":"reno","ak.t":"1732694075","ak.ak":"hOBiQwZUYzCg5VSAfCLimQ==QyPcNbXRsEYrST2JcASsSxc6Pf47pAFtb0GSAryu8JJZH3QmhjU7XdkIAfmkcQroLd3KZzNnxf0C6l15tDOWx0kA0bFzEBrQqxKq0pQqD1kYAdhTI8JlyMfKTJs8bNu/K6nIshl1S2+tFF+xOBWLiCJYv1Dc6hEUAg1TCvBh+QHDjLiQ+hH4yjvsRz+LM+xovq7ka6R2sXimweY4NaaNvFTZbUSHjynLA6T/hhqFYRrz8U7PRCxYuPZq/rgMvrZL+/sG7E5964i4AHpLIlP/UqF0SDj8SoDXrjHXzz/Y0TBp9vlpaPJcKvJAllZ67ziHio5xD9Lj2gIY1icKHBRsKt8+YS26OwVEAEm2yDqKi+rga0wM8BzP7do3laJVhQ3AZg/n5FTpsenLj9ZIte3kywmdOUTZBVtaqBVX87xeBVQ=","ak.pv":"581","ak.dpoabenc":"","ak.tf":i};if(""!==t)_["ak.ruds"]=t;var o={i:!1,av:function(e){var t="http.initiator";if(e&&(!e[t]||"spa_hard"===e[t]))_["ak.feo"]=void 0!==a.aFeoApplied?1:0,BOOMR.addVar(_)},rv:function(){var a=["ak.bpcip","ak.cport","ak.cr","ak.csrc","ak.gh","ak.ipv","ak.m","ak.n","ak.ol","ak.proto","ak.quicv","ak.tlsv","ak.0rtt","ak.0rtt.ed","ak.r","ak.acc","ak.t","ak.tf"];BOOMR.removeVar(a)}};BOOMR.plugins.AK={akVars:_,akDNSPreFetchDomain:n,init:function(){if(!o.i){var a=BOOMR.subscribe;a("before_beacon",o.av,null,null),a("onbeacon",o.rv,null,null),o.i=!0}return this},is_complete:function(){return!0}}}}()}(window);</script></head><body><div id="__next" data-reactroot=""><div class="Layout_mck-c-skipbar__K684J"><a data-component="mdc-c-link" href="#skipToMain" class="mdc-c-button___U4iY2_2734c4f mdc-c-button--primary___Ed-lT_2734c4f mdc-c-button--size-medium"><span class="mdc-c-link__label___Pfqtd_2734c4f">Skip to main content</span></a></div><main class="mck-o-container--outer" data-layer-region="body" role="main" id="skipToMain"><div class="mck-o-container"><div class="mck-o-container--wrapped mck-o-container--mobile-spacing"><div data-component="mdc-c-module-wrapper" data-module-theme="default" data-module-background="transparent" data-module-category="AnchoredHero" class="mdc-u-grid mdc-u-grid-col-md-12"><div class="mdc-u-grid-col-md-start-2 mdc-u-grid-col-md-end-12 mdc-u-grid-col-lg-start-3 mdc-u-grid-col-lg-end-11 MinimalHero_mck-c-minimal-hero__CfxgW "><h1 data-component="mdc-c-heading" class="mdc-c-heading___0fM1W_2734c4f mdc-u-ts-2 mdc-u-align-center"><span>What are AI guardrails?</span></h1><div data-component="mdc-c-description" class="mdc-c-description___SrnQP_2734c4f mdc-u-ts-10 mdc-u-align-center MinimalHero_mck-c-minimal-hero__description__dhVBg"><div><time datetime="2024-11-14T00:00:00Z">November 14, 2024</time> | Article</div></div><section class="mdc-u-grid mdc-u-grid-col-md-12 mck-u-animation-blur-in-400 byline-share-container MinimalHero_mck-c-minimal-hero__authors-and-share-tools__2vQ4V"><div class="mdc-u-grid-col-md-span-12 MinimalHero_mck-c-minimal-hero__share-tools__AIHPk"><div class="mck-u-screen-only"><div data-component="mdc-c-link-container" class="mdc-c-link-container___xefGu_2734c4f mdc-c-link-container--display-column___X0HDD_2734c4f ShareToolBar_mck-c-share-tool-bar__link-container__lW4l5"><div data-component="mdc-c-dropdown-menu" class="mdc-c-drop-down"><button data-component="mdc-c-button" aria-label="" type="button" id="button_id" class="mdc-c-button___U4iY2_2734c4f mdc-c-button--ghost mdc-c-button--size-medium mdc-c-drop-down__rootmenu___yJzvz_2734c4f" aria-expanded="false" aria-haspopup="menu"><span data-component="mdc-c-icon" class="mdc-c-icon___oi7ef_2734c4f mdc-c-icon--default___f-hQM_2734c4f mdc-c-icon--size-md___yi5fA_2734c4f mck-share2-icon"></span></button><div data-component="mdc-c-module-wrapper" data-module-theme="light" data-module-background="white" data-module-category="" class=""></div></div><div data-component="mdc-c-dropdown-menu" class="mdc-c-drop-down"><a data-component="mdc-c-link" href="#/print" class="mdc-c-drop-down__rootmenu___yJzvz_2734c4f mdc-c-link___lBbY1_2734c4f mdc-c-link--secondary___TSFam_2734c4f" data-layer-event-prefix="CTA Link" data-layer-action="click" data-layer-category="share" data-layer-subcategory="print" data-layer-text="Print this article"><span data-component="mdc-c-icon" class="mdc-c-icon___oi7ef_2734c4f mdc-c-icon--default___f-hQM_2734c4f mdc-c-icon--size-md___yi5fA_2734c4f mck-print-icon"></span></a><div data-component="mdc-c-module-wrapper" data-module-theme="light" data-module-background="white" data-module-category="" class=""></div></div><div data-component="mdc-c-dropdown-menu" class="mdc-c-drop-down"><button data-component="mdc-c-button" aria-label="" type="button" id="button_id" class="mdc-c-button___U4iY2_2734c4f mdc-c-button--ghost mdc-c-button--size-medium mdc-c-drop-down__rootmenu___yJzvz_2734c4f" aria-expanded="false" aria-haspopup="menu"><span data-component="mdc-c-icon" class="mdc-c-icon___oi7ef_2734c4f mdc-c-icon--default___f-hQM_2734c4f mdc-c-icon--size-md___yi5fA_2734c4f mck-download-icon"></span></button><div data-component="mdc-c-module-wrapper" data-module-theme="light" data-module-background="white" data-module-category="" class=""></div></div><div data-component="mdc-c-dropdown-menu" class="mdc-c-drop-down"><a data-component="mdc-c-link" href="#/save" class="mdc-c-drop-down__rootmenu___yJzvz_2734c4f mdc-c-link___lBbY1_2734c4f mdc-c-link--secondary___TSFam_2734c4f" data-layer-event-prefix="CTA Link" data-layer-action="click" data-layer-category="article" data-layer-subcategory="save" data-layer-text="Save this article"><span data-component="mdc-c-icon" class="mdc-c-icon___oi7ef_2734c4f mdc-c-icon--default___f-hQM_2734c4f mdc-c-icon--size-md___yi5fA_2734c4f mck-bookmark-icon"></span></a><div data-component="mdc-c-module-wrapper" data-module-theme="light" data-module-background="white" data-module-category="" class=""></div></div></div></div></div></section></div></div></div></div><div class="mck-o-container"><div class="mck-o-container--wrapped mck-o-container--mobile-spacing mdc-u-grid mdc-u-grid-gutter-xxl"><section class="mdc-u-grid mdc-u-grid-col-md-12 mck-u-animation-blur-in-400"><div class="mdc-u-grid-col-md-start-2 mdc-u-grid-col-md-end-12 mdc-u-grid-col-lg-start-3 mdc-u-grid-col-lg-end-11"><div data-component="mdc-c-description" class="mdc-c-description___SrnQP_2734c4f mdc-u-ts-5"><div class="mck-u-links-inline">AI guardrails help ensure that an organization’s AI tools, and their application in the business, reflect the organization’s standards, policies, and values.</div></div></div></section><main data-layer-region="article-body" class="mdc-u-grid mdc-u-grid-gutter-xxl"><div class="mdc-u-grid mdc-u-grid-col-1 mdc-u-grid-col-md-12"><div class="mdc-u-grid-col-md-start-2 mdc-u-grid-col-md-end-12 mdc-u-grid-col-lg-start-3 mdc-u-grid-col-lg-end-11"><div class="mdc-o-content-body mck-u-dropcap"><div data-component="mdc-c-module-wrapper" data-module-theme="default" data-module-background="transparent" data-module-category="" class="mck-o-md-center Image_mck-c-image__fhx5v"><picture data-component="mdc-c-picture" class="Image_mck-c-image__img-container__JusDB"><source media="(min-width: 768px)" srcSet="/~/media/mckinsey/featured%20insights/mckinsey%20explainers/what%20are%20ai%20guardrails/what%20are%20ai%20guardrails_-2152543981-hero-1536x864.jpg?cq=50&mw=1180&cpy=Center"/><img alt="A pair of red and white concrete road barriers aligned on a street against a light blue background. " src="/~/media/mckinsey/featured%20insights/mckinsey%20explainers/what%20are%20ai%20guardrails/what%20are%20ai%20guardrails_-2152543981-hero-1536x864.jpg?cq=50&mw=767&cpy=Center" loading="lazy"/></picture><div class="mck-u-sr-only">A pair of red and white concrete road barriers aligned on a street against a light blue background. </div></div> <div data-component="mdc-c-module-wrapper" data-module-theme="default" data-module-background="transparent" data-module-category="" class="DownloadsSidebar_mck-c-downloads-sidebar__iFmyt mck-o-xs-right-span"><div data-layer-region="downloads-right-rail"><h3 data-component="mdc-c-heading" class="mdc-c-heading___0fM1W_2734c4f mdc-c-heading--title___5qyOB_2734c4f mdc-c-heading--border___K8dj3_2734c4f"></h3><div><div data-component="mdc-c-link-container" class="mdc-c-link-container___xefGu_2734c4f"><a data-component="mdc-c-link" href="#/download/%2F~%2Fmedia%2Fmckinsey%2Ffeatured%20insights%2Fmckinsey%20explainers%2Fwhat%20are%20ai%20guardrails%2Fwhat-are-ai-guardrails.pdf%3FshouldIndex%3Dfalse" class="DownloadsSidebar_mck-c-downloads-sidebar__download-link__fPqFQ mdc-c-link___lBbY1_2734c4f" target="_self" data-layer-event-prefix="Download Link" data-layer-action="click" data-layer-report-type="" data-layer-file-name="what-are-ai-guardrails" data-layer-report-name="what-are-ai-guardrails>"><span data-component="mdc-c-icon" class="mdc-c-icon___oi7ef_2734c4f mdc-c-icon--size-md___yi5fA_2734c4f mck-download-icon"></span><span class="mdc-c-link__label___Pfqtd_2734c4f"> (5 pages)</span></a></div></div></div></div> <p><strong>You know about</strong> guardrails on the highway: barriers along the edge of the road that protect vehicles from veering off course and into danger. With the advent of generative AI (gen AI), the concept of guardrails also applies to systems designed to ensure that a company’s AI tools, especially <a href="/featured-insights/mckinsey-explainers/what-is-deep-learning">large language models</a> (LLMs), work in alignment with organizational standards, policies, and values.</p> <div data-component="mdc-c-module-wrapper" data-module-theme="default" data-module-background="transparent" data-module-category="" class="mck-o-sm-right-span Image_mck-c-image__fhx5v"><div data-component="mdc-c-content-block" class="mdc-c-content-block___7p6Lu_2734c4f"><div data-component="mdc-c-description" class="mdc-c-description___SrnQP_2734c4f mdc-u-ts-9 mdc-u-mt-3 Image_mck-c-image__description__nsDz3"><div><h5>Get to know and directly engage with senior McKinsey experts on AI guardrails</h5> <p><strong><a href="/our-people/lareina-yee">Lareina Yee</a></strong> is a senior partner in McKinsey’s Bay Area office, where <strong><a href="/our-people/roger-roberts">Roger Roberts</a></strong> is a partner; <strong>Mara Pometti</strong> is a consultant in the London office; and <strong>Stephen Xu</strong> is a senior director of product management in the Toronto office.</p></div></div></div><div class="mck-u-sr-only"></div></div> <p>While <a href="/featured-insights/mckinsey-explainers/what-is-generative-ai">gen AI</a> can improve a company’s efficiency, innovation, and competitive advantage, it can also introduce challenges and risks. As companies’ adoption of <a href="/capabilities/quantumblack/our-insights/the-state-of-ai">gen AI increases rapidly</a>, guardrails are crucial to the <a href="/capabilities/quantumblack/how-we-help-clients/generative-ai/responsible-ai-principles">responsible use of AI</a>. Guardrails can identify and remove inaccurate content that’s generated by LLMs, as well as monitor and filter out risky prompts. Such risky content can include security vulnerabilities, hallucinations, toxic or inappropriate content, and misinformation.</p> <!-- --> <p>But just as guardrails on the highway don’t eliminate the risk of injuries or fatalities, AI guardrails don’t guarantee that AI systems will be completely safe, fair, compliant, and ethical. For the best results, companies can implement AI guardrails along with other procedural controls (for example, AI trust frameworks, monitoring and compliance software, testing and evaluation practices), as well as a proper AI operations technology stack, which scales the governance of AI across an organization.</p> <h2>What are the benefits of AI guardrails?</h2> <p>To create the right environment for gen AI innovation and transformation, organizations should ensure that the technology can <a href="/capabilities/risk-and-resilience/our-insights/implementing-generative-ai-with-speed-and-safety">operate safely and responsibly—with AI guardrails playing a critical role</a>. Here are a few benefits that guardrails can offer an organization as it implements AI:</p> <ul> <li><em>Privacy and security.</em> AI systems are susceptible to attacks from malicious actors who exploit vulnerabilities to manipulate AI-generated outcomes. Guardrails can shore up AI systems against such attacks, helping to protect an organization and its customers.</li> <li><em>Regulatory compliance.</em> With <a href="/capabilities/risk-and-resilience/our-insights/as-gen-ai-advances-regulators-and-risk-functions-rush-to-keep-pace">increasing government scrutiny</a> of AI, organizations need to ensure that their AI systems comply with existing and emerging laws and standards. By helping a company maintain its gen AI compliance, guardrails can mitigate the risk of legal penalties and liabilities from the use of these tools.</li> <li><em>Trust.</em> Maintaining trust with customers and the broader public is paramount for organizations. Guardrails enable continuous monitoring and review of AI-generated outputs, which can reduce the risk of errant content being released outside of the company.</li> </ul> <h2>What are the main types of AI guardrails?</h2> <p>Guardrails are grouped according to their purpose and the types of risks they address. (For more information about our methodology for creating guardrails, see sidebar, “What is HyPe?”) McKinsey has developed a taxonomy of guardrails, based on specific risks:</p> <ul> <li><em>Appropriateness</em> guardrails check if the content generated by AI is toxic, harmful, biased, or based on stereotypes and filter out any such inappropriate content before it reaches customers.</li> <li><em>Hallucination</em> guardrails ensure that AI-generated content doesn’t contain information that is factually wrong or misleading.</li> <li><em>Regulatory-compliance</em> guardrails validate that generated content meets regulatory requirements, whether those requirements are general or specific to the industry or use case.</li> <li><em>Alignment</em> guardrails ensure that generated content aligns with user expectations and doesn’t drift away from its main purpose. These guardrails can help maintain brand consistency, for example.</li> <li><em>Validation</em> guardrails check that generated content meets specific criteria: that is, the content contains or does not contain certain information. If a piece of generated content is flagged by a validation guardrail, the content can be funneled into a correction loop to fix the error. Validation should be the last of a series of tasks that guardrails perform. After that, a human validator should review flagged or ambiguous cases that require human reasoning.</li> </ul> <div data-component="mdc-c-module-wrapper" data-module-theme="light" data-module-background="lightest-grey" data-module-category="" data-module-gradient-position="bottom-right" class="mck-c-inline-module-container SideBar_mck-c-sidebar__bgimg-wrapper__Qj4Dt mck-o-sm-left-span SideBar_mck-c-sidebar__sidebar-wrapper__Dpjw2 SideBar_mck-c-sidebar__sidebar-wrapper--istablet__IQ6ii mck-u-screen-only mck-c-module-wrapper" data-layer-region="sidebar"><div class="SideBar_mck-c-sidebar__epoAm mck-o-md-center"><div class="SideBar_mck-c-sidebar__share-icons-wrapper__9gB_c"><div data-component="mdc-c-link-container" class="mdc-c-link-container___xefGu_2734c4f mdc-c-link-container--display-column___X0HDD_2734c4f SideBar_mck-c-sidebar__share-icons___eQy6"><div data-component="mdc-c-dropdown-menu" class="mdc-c-drop-down"><button data-component="mdc-c-button" aria-label="" type="button" id="button_id" class="mdc-c-button___U4iY2_2734c4f mdc-c-button--ghost mdc-c-button--size-medium mdc-c-drop-down__rootmenu___yJzvz_2734c4f" aria-expanded="false" aria-haspopup="menu"><span data-component="mdc-c-icon" class="mdc-c-icon___oi7ef_2734c4f mdc-c-icon--default___f-hQM_2734c4f mdc-c-icon--size-md___yi5fA_2734c4f mck-share2-icon"></span></button><div data-component="mdc-c-module-wrapper" data-module-theme="light" data-module-background="white" data-module-category="" class=""></div></div><button data-component="mdc-c-button" aria-label="Expandable Sidebar" type="button" id="button_id" class="mdc-c-button___U4iY2_2734c4f mdc-c-button--ghost mdc-c-button--size-medium SideBar_mck-c-sidebar__toggle-btn__EL8iE" aria-expanded="false" data-layer-event-prefix="UI Item" data-layer-action="click" data-layer-category="sidebar" data-layer-subcategory="open" data-layer-text="open sidebar"><span data-component="mdc-c-icon" class="mdc-c-icon___oi7ef_2734c4f mdc-c-icon--radial___y3csX_2734c4f mdc-c-icon--size-xxl___cL3ZT_2734c4f mck-plus-no-circle-icon"></span></button></div></div><div class="SideBar_mck-c-sidebar__content-outer__UdWCq"><div class="SideBar_mck-c-sidebar__eyebrow__5GSEq"></div><div class="SideBar_mck-c-sidebar__content__raEwe"><h2 data-component="mdc-c-heading" class="mdc-c-heading___0fM1W_2734c4f mdc-u-ts-3 SideBar_mck-c-sidebar__content-heading__NJekY"><div>What is HyPe?</div></h2><div class="SideBar_mck-c-sidebar__content-description__4p9iI mdc-u-ts-7"><div class="mdc-o-content-body"><p><strong>HyPe is a set</strong> of technology-agnostic, adaptable components developed by McKinsey’s CustomerOne service line and powered by QuantumBlack, AI by McKinsey. Organizations can deploy HyPe components to accelerate hyperpersonalized customer engagements such as marketing messages. As part of the HyPe development process, we held workshops with multiple stakeholders to discuss the risks associated with hyperpersonalized content developed by AI as well as strategies to mitigate these risks. These workshops, along with analysis of the performance of the large language model used during the building and testing phases of HyPe, allowed us to create the guardrail categories described in this article. </p></div></div></div></div></div></div> <p>A variety of open-source libraries have been developed so organizations can add guardrails to their AI systems easily. The machine learning platform Hugging Face has released the Chatbot Guardrails Arena, which stress-tests LLMs and privacy guardrails to prevent leaks of sensitive data. Nvidia has built NeMo Guardrails, an open-source tool kit for adding programmable guardrails to LLM-based applications. Guardrails AI is a similar open-source package. LangChain, a framework for developing applications powered by LLMs, also provides a <a href="https://github.com/langchain-ai/langchain/tree/v0.2/templates/guardrails-output-parser/">guardrails</a> library to help organizations quickly plug guardrails into the sequence of operations. There are also proprietary tools, such as OpenAI’s Moderation, that analyze text generated by AI models to detect and filter out harmful, inappropriate, or unsafe content, according to predefined categories. Microsoft has developed a similar guardrail to monitor chatbot-generated content for Azure, its suite of AI services.</p> <h2>How do guardrails work?</h2> <p>Guardrails are built using a variety of techniques, from rule-based systems to LLMs. Ultimately, though, most guardrails are fully deterministic, meaning the systems always produce the same output for the same input, with no randomness or variability. Generally, guardrails monitor AI systems’ output by performing a range of tasks: for example, classification, semantic validation, detection of personally identifiable information leaks, and identification of harmful content. To perform these tasks, AI guardrails are made up of four interrelated components, each of which plays a crucial role:</p> <ul> <li><em>Checker.</em> The checker scans AI-generated content to detect errors and flag issues, such as offensive language or biased responses. It acts as the first line of defense, identifying potential problems before they can cause harm or violate ethical guidelines.</li> <li><em>Corrector.</em> Once the checker identifies an issue, the corrector refines, corrects, and/or improves the AI’s output as needed. It can correct inaccuracies, remove inappropriate content, and ensure that the response is both precise and aligned with the intended message. The corrector works iteratively, refining the content until it meets the required standards.</li> <li><em>Rail.</em> The rail manages the interaction between the checker and corrector. It runs checks on the content and, if the content fails to meet any standard, triggers the corrector to make adjustments. This process is repeated until the content passes all checks or reaches a predefined correction limit. The rail also logs the processes of the checker and corrector, providing data for further analysis.</li> <li><em>Guard.</em> The guard interacts with all three of the other components, initiating checkers and correctors along with rails, coordinating and managing rails, aggregating the results from rails, and delivering corrected messages.</li> </ul> <div data-component="mdc-c-module-wrapper" data-module-theme="default" data-module-background="transparent" data-module-category="" class="mck-c-disruptor1up mck-o-md-center mck-u-inline-module-border-top mck-u-inline-module-border-bottom mck-u-screen-only" data-layer-region="disruptor-1up"><div class="mdc-u-grid mdc-u-grid-gutter-md mdc-u-grid-col-lg-12 mdc-u-grid-col-md-12 "><div class="mdc-u-grid-col-lg-span-4 mdc-u-grid-col-md-span-5 Disruptor1Up_mck-c-disruptor1up__image___2Gc4"><picture data-component="mdc-c-picture"><img alt="Circular, white maze filled with white semicircles." src="/~/media/mckinsey/featured%20insights/mckinsey%20explainers/mckinseyexplainers-flat-thumb-1536x1536.jpg?cq=50&mh=145&car=16:9&cpy=Center" loading="lazy"/></picture></div><div class="mdc-u-grid-col-lg-span-8 mdc-u-grid-col-md-span-7"><header data-component="mdc-c-header" class="mdc-c-header"><div class="mdc-c-header__block___i1Lg-_2734c4f"><h3 data-component="mdc-c-heading" class="mdc-c-heading___0fM1W_2734c4f"><div>Looking for direct answers to other complex questions?</div></h3></div></header><div data-component="mdc-c-link-container" class="mdc-c-link-container___xefGu_2734c4f mdc-c-link-container--display-column___X0HDD_2734c4f mck-c-disruptor1up__content Disruptor1Up_mck-c-disruptor1up__content--links__VV4lE mdc-u-grid-gutter-md"><a data-component="mdc-c-link" href="/featured-insights/mckinsey-explainers" class="mdc-c-link-cta___NBQVi_2734c4f"><span class="mdc-c-link__label___Pfqtd_2734c4f">Explore the full McKinsey Explainers series</span><span data-component="mdc-c-icon" class="mdc-c-icon___oi7ef_2734c4f mck-link-arrow-right-icon"></span></a></div></div></div></div> <p>When designing guardrails, organizations should ensure that they can be easily integrated with their existing technology stacks and customizable to meet the needs of different use cases.</p> <p>AI agents are also emerging as tools that can function as guardrails. Organizations can use AI agents to automatically check and correct LLM-produced content that has been flagged by guardrails. Early models of AI agents can autonomously monitor, adjust, and regulate AI-generated outputs, as other AI guardrails can do.</p> <h2>How can AI guardrails generate value?</h2> <p>AI guardrails are not only a tool for meeting compliance or ethical requirements; they can also help create a competitive advantage. For one thing, guardrails can help organizations build trust with their customers and avoid costly legal issues. They can also help organizations use AI more responsibly, and thereby attract and retain top talent. </p> <p>To maximize the potential for value creation, organizations can scale their AI guardrails by embedding them into enterprise platforms. <a href="/about-us/overview/alliances-and-acquisitions/iguazio">Iguazio</a> by McKinsey provides AI guardrails in the production environment to help ensure AI governance at scale and reduce the risks of data privacy breaches, bias, hallucinations, and intellectual property infringement.</p> <p>The financial-services company ING offers an example of how an organization can create value with AI guardrails. ING <a href="/industries/financial-services/how-we-help-clients/banking-on-innovation-how-ing-uses-generative-ai-to-put-people-first">developed an AI chatbot</a> with guardrails to ensure accurate and safe customer interactions. The guardrails were applied to filter out sensitive information and potentially risky advice to customers, as well as ensure compliance. Since this was a customer support tool, it was vital from the beginning to design the AI chatbot with guardrails to ensure it provided safe outputs to ING’s customers while also complying with regulatory standards.</p> <h2>How can an organization deploy AI guardrails at scale?</h2> <p>Here are some initial, high-level steps that companies can take:</p> <ul> <li><em>Design the guardrails with multidisciplinary teams.</em> Work with diverse stakeholders, including legal teams, to build guardrails based on the actual risks and effects that might stem from AI.</li> <li><em>Define content quality metrics.</em> These metrics should be tailored to the desired content outputs and based on specific business goals, standards, and regulations. They could include factors such as offensiveness, bias, and alignment with brand guidelines.</li> <li><em>Take a modular approach.</em> Build guardrail components that are reconfigurable for different gen AI use cases and can be easily embedded—and also scaled—in the company’s existing systems.</li> <li><em>Adopt a dynamic approach.</em> Gen AI tools are probabilistic systems that dynamically adjust their outputs based on user-generated inputs. This means that the same input might not always result in exactly the same output, which can sometimes be a problem. An organization should put in place rule-based guardrails with dynamic baselines for a model’s outputs that can change based on different variables.</li> <li><em>Steer with existing regulatory frameworks.</em> Use existing and emerging regulatory, legal, and compliance frameworks, as well as industry best practices, to create “goals” for the guardrails to hit. These can all be used as metrics against which companies can measure their models’ performance.</li> <li><em>Develop new capabilities and roles.</em> Upskill a new generation of practitioners who are accountable for the models’ outcomes and for ensuring AI transparency, governance, and fairness—by, for example, embedding documentation, accountability, and compliance processes into organizations’ ways of working with AI-based tools.</li> </ul> <hr/> <p>The rapid rise of AI has complicated the compliance picture for companies of all stripes working in, and with, technology. Guardrails can help companies get ahead of related risks and create a safer space for gen-AI-related innovation and transformation. For example, organizations could apply AI guardrails to product development, where active safety testing is a critical step. Product development processes, which are typically owned by product leaders or engineers, would need to become more multidisciplinary to incorporate the perspectives of ethicists, as well as leaders in compliance, risk, and operations. While it might seem that all of these steps and changes could slow things down for a company, they are actually designed to help organizations better manage AI-related crises and, hopefully, avert them altogether. Moving forward, we can expect not only new types of AI systems but also new standards for how these systems are developed and operationalized.</p> <p><em>Articles referenced:</em></p> <ul> <li>“<a href="/capabilities/quantumblack/our-insights/the-state-of-ai">The state of AI in early 2024: Gen AI adoption spikes and starts to generate value</a>,” May 30, 2024, <a href="/our-people/alex-singla">Alex Singla</a>, <a href="/our-people/alexander-sukharevsky">Alexander Sukharevsky</a>, <a href="/our-people/lareina-yee">Lareina Yee</a>, and <a href="/our-people/michael-chui">Michael Chui</a>, with <a href="/our-people/bryce-hall">Bryce Hall</a></li> <li>“<a href="/capabilities/risk-and-resilience/our-insights/implementing-generative-ai-with-speed-and-safety">Implementing generative AI with speed and safety</a>,” <em>McKinsey Quarterly,</em> March 13, 2024, Oliver Bevan, <a href="/our-people/michael-chui">Michael Chui</a>, <a href="/our-people/ida-kristensen">Ida Kristensen</a>, Brittany Presten, and <a href="/our-people/lareina-yee">Lareina Yee</a></li> </ul> <div data-component="mdc-c-module-wrapper" data-module-theme="default" data-module-background="transparent" data-module-category="" class="mck-c-disruptor1up mck-o-md-center mck-u-inline-module-border-top mck-u-inline-module-border-bottom mck-u-screen-only" data-layer-region="disruptor-1up"><div class="mdc-u-grid mdc-u-grid-gutter-md mdc-u-grid-col-lg-12 mdc-u-grid-col-md-12 "><div class="mdc-u-grid-col-lg-span-4 mdc-u-grid-col-md-span-5 Disruptor1Up_mck-c-disruptor1up__image___2Gc4"><picture data-component="mdc-c-picture"><img alt="A pair of red and white concrete road barriers aligned on a street against a light blue background. " src="/~/media/mckinsey/featured%20insights/mckinsey%20explainers/what%20are%20ai%20guardrails/what%20are%20ai%20guardrails_-2152543981-thumb-1536x1536.jpg?cq=50&mh=145&car=16:9&cpy=Center" loading="lazy"/></picture></div><div class="mdc-u-grid-col-lg-span-8 mdc-u-grid-col-md-span-7"><header data-component="mdc-c-header" class="mdc-c-header"><div class="mdc-c-header__block___i1Lg-_2734c4f"><h3 data-component="mdc-c-heading" class="mdc-c-heading___0fM1W_2734c4f"><div>Want to know more about AI guardrails?</div></h3><div data-component="mdc-c-description" class="mdc-c-description___SrnQP_2734c4f mdc-u-ts-7"><p><div><p style="text-align: left;"><a class="btn btn-fill" href="/capabilities/mckinsey-digital/how-we-help-clients">Talk to us</a></p></div></p></div></div></header></div></div></div></div><div class="container-placeholder"></div></div></div></main></div><div data-component="mdc-c-module-wrapper" data-module-theme="light" data-module-background="lightest-grey" data-module-category="StandalonePromo" class="RelatedArticle_mck-c-article-related__GGA76 mck-u-screen-only" data-layer-region="related-articles"><div class="mdc-o-container__wrapper is-wrapped mdc-u-spaced-mobile"><h5 data-component="mdc-c-heading" class="mdc-c-heading___0fM1W_2734c4f mdc-c-heading--title___5qyOB_2734c4f mdc-u-align-center">Related Articles</h5><div class="mdc-u-grid mdc-u-grid-col-md-3 RelatedArticle_items-container__s2uD0"><div class="mdc-u-grid mdc-u-grid-gutter-lg Card_card__diA2r Card_hover-effect__RGb9b"><div class="Card_wrapper-image__8b4P6"><a data-component="mdc-c-link" href="/featured-insights/mckinsey-explainers/what-is-retrieval-augmented-generation-rag" class="mdc-c-link___lBbY1_2734c4f"><picture data-component="mdc-c-picture" class="Card_wrapper-image__8b4P6"><style>.picture-uniqueKey-what-0 { aspect-ratio: 16/9 }</style><img alt="A golden outline of a speech bubble is filled with a jumble of colorful, balloon-like spheres. The spheres vary in size, color, and texture, and the image is set against a light-gray background." class="picture-uniqueKey-what-0" src="/~/media/mckinsey/featured%20insights/mckinsey%20explainers/what%20is%20retrieval%20augmented%20generation%20rag/what%20is%20rag_-1778431906-thumb-1536x1536.jpg?cq=50&mw=767&car=16:9&cpy=Center" loading="lazy"/></picture></a></div><div class="Card_wrapper-text__U6Y3k"><div data-component="mdc-c-content-block" class="mdc-c-content-block___7p6Lu_2734c4f Card_content-block__pF6Z1"><span>Article</span><h6 data-component="mdc-c-heading" class="mdc-c-heading___0fM1W_2734c4f mdc-u-ts-6"><a data-component="mdc-c-link" href="/featured-insights/mckinsey-explainers/what-is-retrieval-augmented-generation-rag" class="mdc-c-link-heading___Zggl8_2734c4f mdc-c-link___lBbY1_2734c4f"><div>What is retrieval-augmented generation (RAG)?</div></a></h6></div></div></div><div class="mdc-u-grid mdc-u-grid-gutter-lg Card_card__diA2r Card_hover-effect__RGb9b"><div class="Card_wrapper-image__8b4P6"><a data-component="mdc-c-link" href="/featured-insights/mckinsey-explainers/whats-the-future-of-ai" class="mdc-c-link___lBbY1_2734c4f"><picture data-component="mdc-c-picture" class="Card_wrapper-image__8b4P6"><style>.picture-uniqueKey-what’-0 { aspect-ratio: 16/9 }</style><img alt="Conceptual illustration of 7 glasslike panels floating over a grid. The panels transition from dark to light blue and 2 pink lines weave past the panels and pink dots float around the grid." class="picture-uniqueKey-what’-0" src="/~/media/mckinsey/featured%20insights/mckinsey%20explainers/whats%20the%20future%20of%20ai/whats-the-future-of-ai-thumb-1536x1536.jpg?cq=50&mw=767&car=16:9&cpy=Center" loading="lazy"/></picture></a></div><div class="Card_wrapper-text__U6Y3k"><div data-component="mdc-c-content-block" class="mdc-c-content-block___7p6Lu_2734c4f Card_content-block__pF6Z1"><span>Article</span><h6 data-component="mdc-c-heading" class="mdc-c-heading___0fM1W_2734c4f mdc-u-ts-6"><a data-component="mdc-c-link" href="/featured-insights/mckinsey-explainers/whats-the-future-of-ai" class="mdc-c-link-heading___Zggl8_2734c4f mdc-c-link___lBbY1_2734c4f"><div>What’s the future of AI?</div></a></h6></div></div></div><div class="mdc-u-grid mdc-u-grid-gutter-lg Card_card__diA2r Card_hover-effect__RGb9b"><div class="Card_wrapper-image__8b4P6"><a data-component="mdc-c-link" href="/featured-insights/mckinsey-explainers/what-is-artificial-general-intelligence-agi" class="mdc-c-link___lBbY1_2734c4f"><picture data-component="mdc-c-picture" class="Card_wrapper-image__8b4P6"><style>.picture-uniqueKey-what-0 { aspect-ratio: 16/9 }</style><img alt="A profile of a 3d head made of concrete that is sliced in half creating two separate parts. Pink neon binary numbers travel from one half of the a head to the other by a stone bridge that connects the two parts." class="picture-uniqueKey-what-0" src="/~/media/mckinsey/featured%20insights/mckinsey%20explainers/what%20is%20artificial%20general%20intelligence%20agi/agi-1622583937-thumb-1536x1536.jpg?cq=50&mw=767&car=16:9&cpy=Center" loading="lazy"/></picture></a></div><div class="Card_wrapper-text__U6Y3k"><div data-component="mdc-c-content-block" class="mdc-c-content-block___7p6Lu_2734c4f Card_content-block__pF6Z1"><span>Article</span><h6 data-component="mdc-c-heading" class="mdc-c-heading___0fM1W_2734c4f mdc-u-ts-6"><a data-component="mdc-c-link" href="/featured-insights/mckinsey-explainers/what-is-artificial-general-intelligence-agi" class="mdc-c-link-heading___Zggl8_2734c4f mdc-c-link___lBbY1_2734c4f"><div>What is artificial general intelligence (AGI)?</div></a></h6></div></div></div></div></div></div></div></main></div><script id="__NEXT_DATA__" type="application/json">{"props":{"pageProps":{"locale":"en","dictionary":{},"sitecoreContext":{"route":{"name":"What are AI guardrails","displayName":"What are AI guardrails","fields":null,"databaseName":"web","deviceId":"fe5d7fdf-89c0-4d99-9aa3-b5fbd009c9f3","itemId":"3899126a-e938-4641-bb09-fc15c3868982","itemLanguage":"en","itemVersion":1,"layoutId":"ae753eb4-a035-40b4-83bf-4b4438df6742","templateId":"683910db-02ba-40ba-92e7-726c880160a9","templateName":"ArticleJSS","placeholders":{"jss-main":[{"uid":"232bb7e9-289f-492d-a916-2b6185e44a84","componentName":"ArticleTemplate","dataSource":"","fields":{"data":{"articleTemplate":{"title":{"jsonValue":{"value":"What are AI guardrails?"}},"sEOTitle":{"value":""},"description":{"jsonValue":{"value":"AI guardrails help ensure that an organization\u0026rsquo;s AI tools, and their application in the business, reflect the organization\u0026rsquo;s standards, policies, and values."}},"sEODescription":{"value":""},"displayDate":{"jsonValue":{"value":"2024-11-14T00:00:00Z"}},"body":{"value":"[[Image 1]]\n[[DownloadsSidebar]]\n\u003cp\u003e\u003cstrong\u003eYou know about\u003c/strong\u003e guardrails on the highway: barriers along the edge of the road that protect vehicles from veering off course and into danger. With the advent of generative AI (gen AI), the concept of guardrails also applies to systems designed to ensure that a company\u0026rsquo;s AI tools, especially \u003ca href=\"/featured-insights/mckinsey-explainers/what-is-deep-learning\"\u003elarge language models\u003c/a\u003e\u0026nbsp;(LLMs), work in alignment with organizational standards, policies, and values.\u003c/p\u003e\n[[Image get to know]]\n\u003cp\u003eWhile \u003ca href=\"/featured-insights/mckinsey-explainers/what-is-generative-ai\"\u003egen AI\u003c/a\u003e\u0026nbsp;can improve a company\u0026rsquo;s efficiency, innovation, and competitive advantage, it can also introduce challenges and risks. As companies\u0026rsquo; adoption of \u003ca href=\"/capabilities/quantumblack/our-insights/the-state-of-ai\"\u003egen AI increases rapidly\u003c/a\u003e, guardrails are crucial to the \u003ca href=\"/capabilities/quantumblack/how-we-help-clients/generative-ai/responsible-ai-principles\"\u003eresponsible use of AI\u003c/a\u003e. Guardrails can identify and remove inaccurate content that\u0026rsquo;s generated by LLMs, as well as monitor and filter out risky prompts. Such risky content can include security vulnerabilities, hallucinations, toxic or inappropriate content, and misinformation.\u003c/p\u003e\n[[MostPopularArticles 5]]\n\u003cp\u003eBut just as guardrails on the highway don\u0026rsquo;t eliminate the risk of injuries or fatalities, AI guardrails don\u0026rsquo;t guarantee that AI systems will be completely safe, fair, compliant, and ethical. For the best results, companies can implement AI guardrails along with other procedural controls (for example, AI trust frameworks, monitoring and compliance software, testing and evaluation practices), as well as a proper AI operations technology stack, which scales the governance of AI across an organization.\u003c/p\u003e\n\u003ch2\u003eWhat are the benefits of AI guardrails?\u003c/h2\u003e\n\u003cp\u003eTo create the right environment for gen AI innovation and transformation, organizations should ensure that the technology can \u003ca href=\"/capabilities/risk-and-resilience/our-insights/implementing-generative-ai-with-speed-and-safety\"\u003eoperate safely and responsibly\u0026mdash;with AI guardrails playing a critical role\u003c/a\u003e. Here are a few benefits that guardrails can offer an organization as it implements AI:\u003c/p\u003e\n\u003cul\u003e\n \u003cli\u003e\u003cem\u003ePrivacy and security.\u003c/em\u003e AI systems are susceptible to attacks from malicious actors who exploit vulnerabilities to manipulate AI-generated outcomes. Guardrails can shore up AI systems against such attacks, helping to protect an organization and its customers.\u003c/li\u003e\n \u003cli\u003e\u003cem\u003eRegulatory compliance.\u003c/em\u003e With \u003ca href=\"/capabilities/risk-and-resilience/our-insights/as-gen-ai-advances-regulators-and-risk-functions-rush-to-keep-pace\"\u003eincreasing government scrutiny\u003c/a\u003e\u0026nbsp;of AI, organizations need to ensure that their AI systems comply with existing and emerging laws and standards. By helping a company maintain its gen AI compliance, guardrails can mitigate the risk of legal penalties and liabilities from the use of these tools.\u003c/li\u003e\n \u003cli\u003e\u003cem\u003eTrust.\u003c/em\u003e Maintaining trust with customers and the broader public is paramount for organizations. Guardrails enable continuous monitoring and review of AI-generated outputs, which can reduce the risk of errant content being released outside of the company.\u003c/li\u003e\n\u003c/ul\u003e\n\u003ch2\u003eWhat are the main types of AI guardrails?\u003c/h2\u003e\n\u003cp\u003eGuardrails are grouped according to their purpose and the types of risks they address. (For more information about our methodology for creating guardrails, see sidebar, \u0026ldquo;What is HyPe?\u0026rdquo;) McKinsey has developed a taxonomy of guardrails, based on specific risks:\u003c/p\u003e\n\u003cul\u003e\n \u003cli\u003e\u003cem\u003eAppropriateness\u003c/em\u003e guardrails check if the content generated by AI is toxic, harmful, biased, or based on stereotypes and filter out any such inappropriate content before it reaches customers.\u003c/li\u003e\n \u003cli\u003e\u003cem\u003eHallucination\u003c/em\u003e guardrails ensure that AI-generated content doesn\u0026rsquo;t contain information that is factually wrong or misleading.\u003c/li\u003e\n \u003cli\u003e\u003cem\u003eRegulatory-compliance\u003c/em\u003e guardrails validate that generated content meets regulatory requirements, whether those requirements are general or specific to the industry or use case.\u003c/li\u003e\n \u003cli\u003e\u003cem\u003eAlignment\u003c/em\u003e guardrails ensure that generated content aligns with user expectations and doesn\u0026rsquo;t drift away from its main purpose. These guardrails can help maintain brand consistency, for example.\u003c/li\u003e\n \u003cli\u003e\u003cem\u003eValidation\u003c/em\u003e guardrails check that generated content meets specific criteria: that is, the content contains or does not contain certain information. If a piece of generated content is flagged by a validation guardrail, the content can be funneled into a correction loop to fix the error. Validation should be the last of a series of tasks that guardrails perform. After that, a human validator should review flagged or ambiguous cases that require human reasoning.\u003c/li\u003e\n\u003c/ul\u003e\n[[Sidebar Hype]]\n\u003cp\u003eA variety of open-source libraries have been developed so organizations can add guardrails to their AI systems easily. The machine learning platform Hugging Face has released the Chatbot Guardrails Arena, which stress-tests LLMs and privacy guardrails to prevent leaks of sensitive data. Nvidia has built NeMo Guardrails, an open-source tool kit for adding programmable guardrails to LLM-based applications. Guardrails AI is a similar open-source package. LangChain, a framework for developing applications powered by LLMs, also provides a \u003ca href=\"https://github.com/langchain-ai/langchain/tree/v0.2/templates/guardrails-output-parser/\"\u003eguardrails\u003c/a\u003e library to help organizations quickly plug guardrails into the sequence of operations. There are also proprietary tools, such as OpenAI\u0026rsquo;s Moderation, that analyze text generated by AI models to detect and filter out harmful, inappropriate, or unsafe content, according to predefined categories. Microsoft has developed a similar guardrail to monitor chatbot-generated content for Azure, its suite of AI services.\u003c/p\u003e\n\u003ch2\u003eHow do guardrails work?\u003c/h2\u003e\n\u003cp\u003eGuardrails are built using a variety of techniques, from rule-based systems to LLMs. Ultimately, though, most guardrails are fully deterministic, meaning the systems always produce the same output for the same input, with no randomness or variability. Generally, guardrails monitor AI systems\u0026rsquo; output by performing a range of tasks: for example, classification, semantic validation, detection of personally identifiable information leaks, and identification of harmful content. To perform these tasks, AI guardrails are made up of four interrelated components, each of which plays a crucial role:\u003c/p\u003e\n\u003cul\u003e\n \u003cli\u003e\u003cem\u003eChecker.\u003c/em\u003e The checker scans AI-generated content to detect errors and flag issues, such as offensive language or biased responses. It acts as the first line of defense, identifying potential problems before they can cause harm or violate ethical guidelines.\u003c/li\u003e\n \u003cli\u003e\u003cem\u003eCorrector.\u003c/em\u003e Once the checker identifies an issue, the corrector refines, corrects, and/or improves the AI\u0026rsquo;s output as needed. It can correct inaccuracies, remove inappropriate content, and ensure that the response is both precise and aligned with the intended message. The corrector works iteratively, refining the content until it meets the required standards.\u003c/li\u003e\n \u003cli\u003e\u003cem\u003eRail.\u003c/em\u003e The rail manages the interaction between the checker and corrector. It runs checks on the content and, if the content fails to meet any standard, triggers the corrector to make adjustments. This process is repeated until the content passes all checks or reaches a predefined correction limit. The rail also logs the processes of the checker and corrector, providing data for further analysis.\u003c/li\u003e\n \u003cli\u003e\u003cem\u003eGuard.\u003c/em\u003e The guard interacts with all three of the other components, initiating checkers and correctors along with rails, coordinating and managing rails, aggregating the results from rails, and delivering corrected messages.\u003c/li\u003e\n\u003c/ul\u003e\n[[Disruptor1Up mckinseyexplainers]]\n\u003cp\u003eWhen designing guardrails, organizations should ensure that they can be easily integrated with their existing technology stacks and customizable to meet the needs of different use cases.\u003c/p\u003e\n\u003cp\u003eAI agents are also emerging as tools that can function as guardrails. Organizations can use AI agents to automatically check and correct LLM-produced content that has been flagged by guardrails. Early models of AI agents can autonomously monitor, adjust, and regulate AI-generated outputs, as other AI guardrails can do.\u003c/p\u003e\n\u003ch2\u003eHow can AI guardrails generate value?\u003c/h2\u003e\n\u003cp\u003eAI guardrails are not only a tool for meeting compliance or ethical requirements; they can also help create a competitive advantage. For one thing, guardrails can help organizations build trust with their customers and avoid costly legal issues. They can also help organizations use AI more responsibly, and thereby attract and retain top talent. \u003c/p\u003e\n\u003cp\u003eTo maximize the potential for value creation, organizations can scale their AI guardrails by embedding them into enterprise platforms. \u003ca href=\"/about-us/overview/alliances-and-acquisitions/iguazio\"\u003eIguazio\u003c/a\u003e\u0026nbsp;by McKinsey provides AI guardrails in the production environment to help ensure AI governance at scale and reduce the risks of data privacy breaches, bias, hallucinations, and intellectual property infringement.\u003c/p\u003e\n\u003cp\u003eThe financial-services company ING offers an example of how an organization can create value with AI guardrails. ING \u003ca href=\"/industries/financial-services/how-we-help-clients/banking-on-innovation-how-ing-uses-generative-ai-to-put-people-first\"\u003edeveloped an AI chatbot\u003c/a\u003e\u0026nbsp;with guardrails to ensure accurate and safe customer interactions. The guardrails were applied to filter out sensitive information and potentially risky advice to customers, as well as ensure compliance. Since this was a customer support tool, it was vital from the beginning to design the AI chatbot with guardrails to ensure it provided safe outputs to ING\u0026rsquo;s customers while also complying with regulatory standards.\u003c/p\u003e\n\u003ch2\u003eHow can an organization deploy AI guardrails at scale?\u003c/h2\u003e\n\u003cp\u003eHere are some initial, high-level steps that companies can take:\u003c/p\u003e\n\u003cul\u003e\n \u003cli\u003e\u003cem\u003eDesign the guardrails with multidisciplinary teams.\u003c/em\u003e Work with diverse stakeholders, including legal teams, to build guardrails based on the actual risks and effects that might stem from AI.\u003c/li\u003e\n \u003cli\u003e\u003cem\u003eDefine content quality metrics.\u003c/em\u003e These metrics should be tailored to the desired content outputs and based on specific business goals, standards, and regulations. They could include factors such as offensiveness, bias, and alignment with brand guidelines.\u003c/li\u003e\n \u003cli\u003e\u003cem\u003eTake a modular approach.\u003c/em\u003e Build guardrail components that are reconfigurable for different gen AI use cases and can be easily embedded\u0026mdash;and also scaled\u0026mdash;in the company\u0026rsquo;s existing systems.\u003c/li\u003e\n \u003cli\u003e\u003cem\u003eAdopt a dynamic approach.\u003c/em\u003e Gen AI tools are probabilistic systems that dynamically adjust their outputs based on user-generated inputs. This means that the same input might not always result in exactly the same output, which can sometimes be a problem. An organization should put in place rule-based guardrails with dynamic baselines for a model\u0026rsquo;s outputs that can change based on different variables.\u003c/li\u003e\n \u003cli\u003e\u003cem\u003eSteer with existing regulatory frameworks.\u003c/em\u003e Use existing and emerging regulatory, legal, and compliance frameworks, as well as industry best practices, to create \u0026ldquo;goals\u0026rdquo; for the guardrails to hit. These can all be used as metrics against which companies can measure their models\u0026rsquo; performance.\u003c/li\u003e\n \u003cli\u003e\u003cem\u003eDevelop new capabilities and roles.\u003c/em\u003e Upskill a new generation of practitioners who are accountable for the models\u0026rsquo; outcomes and for ensuring AI transparency, governance, and fairness\u0026mdash;by, for example, embedding documentation, accountability, and compliance processes into organizations\u0026rsquo; ways of working with AI-based tools.\u003c/li\u003e\n\u003c/ul\u003e\n\u003chr /\u003e\n\u003cp\u003eThe rapid rise of AI has complicated the compliance picture for companies of all stripes working in, and with, technology. Guardrails can help companies get ahead of related risks and create a safer space for gen-AI-related innovation and transformation. For example, organizations could apply AI guardrails to product development, where active safety testing is a critical step. Product development processes, which are typically owned by product leaders or engineers, would need to become more multidisciplinary to incorporate the perspectives of ethicists, as well as leaders in compliance, risk, and operations. While it might seem that all of these steps and changes could slow things down for a company, they are actually designed to help organizations better manage AI-related crises and, hopefully, avert them altogether. Moving forward, we can expect not only new types of AI systems but also new standards for how these systems are developed and operationalized.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eArticles referenced:\u003c/em\u003e\u003c/p\u003e\n\u003cul\u003e\n \u003cli\u003e\u0026ldquo;\u003ca href=\"/capabilities/quantumblack/our-insights/the-state-of-ai\"\u003eThe state of AI in early 2024: Gen AI adoption spikes and starts to generate value\u003c/a\u003e,\u0026rdquo; May 30, 2024, \u003ca href=\"/our-people/alex-singla\"\u003eAlex Singla\u003c/a\u003e, \u003ca href=\"/our-people/alexander-sukharevsky\"\u003eAlexander Sukharevsky\u003c/a\u003e, \u003ca href=\"/our-people/lareina-yee\"\u003eLareina Yee\u003c/a\u003e, and \u003ca href=\"/our-people/michael-chui\"\u003eMichael Chui\u003c/a\u003e, with \u003ca href=\"/our-people/bryce-hall\"\u003eBryce Hall\u003c/a\u003e\u003c/li\u003e\n \u003cli\u003e\u0026ldquo;\u003ca href=\"/capabilities/risk-and-resilience/our-insights/implementing-generative-ai-with-speed-and-safety\"\u003eImplementing generative AI with speed and safety\u003c/a\u003e,\u0026rdquo; \u003cem\u003eMcKinsey Quarterly,\u003c/em\u003e March 13, 2024, Oliver Bevan, \u003ca href=\"/our-people/michael-chui\"\u003eMichael Chui\u003c/a\u003e, \u003ca href=\"/our-people/ida-kristensen\"\u003eIda Kristensen\u003c/a\u003e, Brittany Presten, and \u003ca href=\"/our-people/lareina-yee\"\u003eLareina Yee\u003c/a\u003e\u003c/li\u003e\n\u003c/ul\u003e\n[[Disruptor1Up connect]]"},"isFullScreenInteractive":{"boolValue":false},"hideStickySocialShareBar":{"boolValue":false},"desktopID":{"value":""},"mobileID":{"value":""},"desktopURL":{"value":""},"mobileURL":{"value":""},"desktopPaddingPercentage":{"value":""},"mobilePaddingPercentage":{"value":""},"desktopOverrideHeight":{"value":""},"mobileOverrideHeight":{"value":""},"cerosOembedURL":{"value":""},"cerosRenderMode":{"targetItem":null},"cerosBackgroundColor":{"targetItem":null},"hideByLine":{"boolValue":false},"tableOfContentsTitle":{"value":"TABLE OF CONTENTS"},"accessStatus":{"targetItem":{"key":{"value":"RegisteredUsers"},"value":{"value":"Registered Users"}}},"articleType":{"targetItem":{"displayName":"Article"}},"hasSpecialReport":{"boolValue":false},"contentType":{"targetItem":{"displayName":"Article"}},"sourcePublication":{"targetItem":null},"externalPublication":{"value":""},"mobileReady":{"boolValue":true},"forClientsOnly":{"boolValue":false},"excludeFromClientLink":{"boolValue":false},"originalPublishDate":{"jsonValue":{"value":"2024-11-14T10:00:00Z"}},"footnotes":{"value":""},"contributoryPractice":{"targetItems":[]},"aboutTheAuthors":{"value":""},"authors":{"targetItems":[]},"nonPartnerAuthors":{"targetItems":[]},"interactiveToUse":{"targetItem":null},"enableArticleComponents":{"boolValue":false},"relatedArticles":{"targetItems":[{"sourcePublication":{"targetItem":null},"publicationSource":null,"externalPublication":{"value":""},"title":{"value":"What is retrieval-augmented generation (RAG)?"},"url":{"path":"/featured-insights/mckinsey-explainers/what-is-retrieval-augmented-generation-rag"},"eyebrow":{"targetItem":{"name":"Article"}},"articleType":{"targetItem":{"name":"Article"}},"contentType":{"targetItem":{"name":"Article"}},"description":{"value":"Retrieval-augmented generation, or RAG, is a process applied to large language models to make their outputs more relevant for the end user."},"standardImage":{"src":"/~/media/mckinsey/featured insights/mckinsey explainers/what is retrieval augmented generation rag/what is rag_-1778431906-thumb-1536x1536.jpg","alt":"A golden outline of a speech bubble is filled with a jumble of colorful, balloon-like spheres. The spheres vary in size, color, and texture, and the image is set against a light-gray background."},"heroImage":null,"thumbnailImage":null},{"sourcePublication":{"targetItem":null},"publicationSource":null,"externalPublication":{"value":""},"title":{"value":"What’s the future of AI?"},"url":{"path":"/featured-insights/mckinsey-explainers/whats-the-future-of-ai"},"eyebrow":{"targetItem":{"name":"Article"}},"articleType":{"targetItem":{"name":"Article"}},"contentType":{"targetItem":{"name":"Article"}},"description":{"value":"AI is here to stay. To outcompete in the future, organizations and individuals alike need to get familiar fast. This series of \u003cem\u003eMcKinsey Explainers\u003c/em\u003e dives deep into the seven technologies that are already shaping the years to come."},"standardImage":{"src":"/~/media/mckinsey/featured insights/mckinsey explainers/whats the future of ai/whats-the-future-of-ai-thumb-1536x1536.jpg","alt":"Conceptual illustration of 7 glasslike panels floating over a grid. The panels transition from dark to light blue and 2 pink lines weave past the panels and pink dots float around the grid."},"heroImage":null,"thumbnailImage":null},{"sourcePublication":{"targetItem":null},"publicationSource":null,"externalPublication":{"value":""},"title":{"value":"What is artificial general intelligence (AGI)?"},"url":{"path":"/featured-insights/mckinsey-explainers/what-is-artificial-general-intelligence-agi"},"eyebrow":{"targetItem":{"name":"Article"}},"articleType":{"targetItem":{"name":"Article"}},"contentType":{"targetItem":{"name":"Article"}},"description":{"value":"Artificial general intelligence (AGI) is a theoretical AI system with capabilities that rival those of a human. Many researchers believe we are still decades, if not centuries, away from achieving AGI."},"standardImage":{"src":"/~/media/mckinsey/featured insights/mckinsey explainers/what is artificial general intelligence agi/agi-1622583937-thumb-1536x1536.jpg","alt":"A profile of a 3d head made of concrete that is sliced in half creating two separate parts. Pink neon binary numbers travel from one half of the a head to the other by a stone bridge that connects the two parts."},"heroImage":null,"thumbnailImage":null}]},"useEnhancedAuthors":{"boolValue":false},"acknowledgements":{"value":""},"showSocialShareFooter":{"boolValue":false},"template":{"id":"683910db-02ba-40ba-92e7-726c880160a9"},"hideFromSearchEngines":{"boolValue":false},"heroFiftyFiftyBackground":{"targetItem":null},"backgroundColor":{"targetItem":null},"gradientDirection":{"targetItem":{"key":{"value":"Bottom Right"},"value":{"value":"bottom-right"}}},"hideStickySubscriptionBar":{"value":false},"enableRegWall":{"boolValue":false},"timer":{"value":"8"},"regWallHeading":{"value":""},"showGoToHomeLink":{"boolValue":false},"regWallDescription":{"value":""},"isInsightsStorePage":{"boolValue":false},"enableRatingsForArticleBody":{"boolValue":false},"enableRatingsForMostPopular":{"boolValue":false},"selectedModalSubscriptions":{"targetItems":[]},"link1":{"queryString":"","className":"","anchor":"","linkType":"internal","url":"","text":"","targetItem":null},"link2":{"queryString":"","className":"","anchor":"","linkType":"internal","url":"","text":"","targetItem":null},"link3":{"queryString":"","className":"","anchor":"","linkType":"internal","url":"","text":"","targetItem":null},"link4":{"queryString":"","className":"","anchor":"","linkType":"internal","url":"","text":"","targetItem":null},"link5":{"queryString":"","className":"","anchor":"","linkType":"internal","url":"","text":"","targetItem":null},"fullReportPDF":{"name":"Full Report PDF","src":null,"displayName":null,"title":null,"keywords":null,"description":null,"extension":null,"mimeType":null,"size":null,"pageCount":null,"thumbnailImageSrc":""},"fullReportPDFDisplayName":{"value":""},"appendixPDF":{"name":"Appendix PDF","src":null,"displayName":null,"title":null,"keywords":null,"description":null,"extension":null,"mimeType":null,"size":null,"pageCount":null,"thumbnailImageSrc":""},"appendixPDFDisplayName":{"value":""},"articlePDF":{"name":"Article PDF","src":"/~/media/mckinsey/featured insights/mckinsey explainers/what are ai guardrails/what-are-ai-guardrails.pdf","displayName":"What-are-AI-guardrails","title":"What are AI guardrails?","keywords":"","description":"","extension":"pdf","mimeType":"application/pdf","size":333968,"pageCount":"5","thumbnailImageSrc":"/~/media/mckinsey/featured insights/mckinsey explainers/what are ai guardrails/what-are-ai-guardrails_thumbnail.jpeg"},"briefingNotePDF":{"name":"Briefing Note PDF","src":null,"displayName":null,"title":null,"keywords":null,"description":null,"extension":null,"mimeType":null,"size":null,"pageCount":null,"thumbnailImageSrc":""},"discussionPapersPDF":{"name":"Discussion Papers PDF","src":null,"displayName":null,"title":null,"keywords":null,"description":null,"extension":null,"mimeType":null,"size":null,"pageCount":null,"thumbnailImageSrc":""},"executiveSummaryPDF":{"name":"Executive Summary PDF","src":null,"displayName":null,"title":null,"keywords":null,"description":null,"extension":null,"mimeType":null,"size":null,"pageCount":null,"thumbnailImageSrc":""},"executiveSummaryPDFDisplayName":{"value":""},"researchPreviewPDF":{"name":"Research Preview PDF","src":null,"displayName":null,"title":null,"keywords":null,"description":null,"extension":null,"mimeType":null,"size":null,"pageCount":null,"thumbnailImageSrc":""},"heroImage":{"src":"/~/media/mckinsey/featured insights/mckinsey explainers/what are ai guardrails/what are ai guardrails_-2152543981-hero-1536x864.jpg","alt":"A pair of red and white concrete road barriers aligned on a street against a light blue background. "},"standardImage":{"src":"/~/media/mckinsey/featured insights/mckinsey explainers/what are ai guardrails/what are ai guardrails_-2152543981-thumb-1536x1536.jpg","alt":"A pair of red and white concrete road barriers aligned on a street against a light blue background. "},"heroImageTopOffset":{"value":""},"headerOverlayOpacity":{"targetItem":null},"heroType":{"targetItem":{"displayName":"Minimal Hero"}},"photoOverlayOpacity":{"targetItem":null},"hideHero":{"boolValue":false},"renderStackedHeroLayoutForMobile":{"boolValue":false},"articleHeroVideoId":{"value":""},"podcastHeroOmnyAudioID":{"value":""},"heroBespokeInteractiveID":{"value":""},"bespokeHeroRenderMode":{"targetItem":null},"audio":{"results":[]},"disruptor1up":{"results":[{"title":{"value":"Want to know more about AI guardrails?"},"description":{"value":"\u003cp style=\"text-align: left;\"\u003e\u003ca class=\"btn btn-fill\" href=\"/capabilities/mckinsey-digital/how-we-help-clients\"\u003eTalk to us\u003c/a\u003e\u003c/p\u003e"},"image":{"src":"/~/media/mckinsey/featured insights/mckinsey explainers/what are ai guardrails/what are ai guardrails_-2152543981-thumb-1536x1536.jpg","alt":"A pair of red and white concrete road barriers aligned on a street against a light blue background. "},"moreLinks":{"jsonValue":{"value":[]}},"mediaID":{"value":"connect"},"template":{"name":"Disruptor 1Up"}},{"mediaID":{"value":"cop"},"title":{"value":"Explore COP29 with McKinsey"},"description":{"value":"Join our series of dynamic virtual events during COP29. Discover new research, practical strategies, and collaborations across sectors to move climate action beyond why to how."},"image":{"src":"/~/media/mckinsey/business functions/sustainability/how we help clients/cop/cop29/cop_29_thumb.jpg","alt":"McKinsey at COP29"},"moreLinks":{"jsonValue":{"value":[{"CssClass":"","LinkType":"external","Target":"","Text":"Register","Title":"","Url":"https://mckinsey.cventevents.com/Q2MVE1?rt=RrQJ1iBSfEiqSK7N9VAGCw\u0026RefId=cop"}]}}},{"mediaID":{"value":"attheedge"},"title":{"value":"Subscribe to the \u003cem\u003eAt the Edge\u003c/em\u003e podcast"},"description":{"value":""},"image":{"src":null,"alt":""},"moreLinks":{"jsonValue":{"value":[{"CssClass":"btn btn-fill","LinkType":"external","Target":"","Text":"Apple Podcasts","Title":"","Url":"https://podcasts.apple.com/us/podcast/at-the-edge/id1612870236"},{"CssClass":"btn btn-fill","LinkType":"external","Target":"","Text":"Spotify","Title":"","Url":"https://open.spotify.com/show/3SnUcezwM20nKnzqvUl6iF"}]}}},{"mediaID":{"value":"authortalks"},"title":{"value":"Visit \u003cem\u003e\u003ca href=\"/featured-insights/mckinsey-on-books/author-talks\"\u003eAuthor Talks\u003c/a\u003e\u003c/em\u003e to see the full series."},"description":{"value":""},"image":{"src":"/~/media/mckinsey/featured insights/mckinsey books/author talks/authortalks_1536x1536_v1.png","alt":"Author Talks"},"moreLinks":{"jsonValue":{"value":[{"CssClass":"","LinkType":"internal","Target":"","Text":"Learn more","Title":"","Url":"/featured-insights/mckinsey-on-books/author-talks"}]}}},{"mediaID":{"value":"COVID"},"title":{"value":"Explore McKinsey’s ongoing coverage of the pandemic"},"description":{"value":""},"image":{"src":"/~/media/mckinsey/business functions/risk/our insights/covid 19 implications for business/implications-briefing-note-4-standard-1536x1536.jpg","alt":"COVID-19: Implications for business"},"moreLinks":{"jsonValue":{"value":[{"CssClass":"","LinkType":"internal","Target":"","Text":"Read more about the business impact of coronavirus and how organizations can respond","Title":"","Url":"/featured-insights/coronavirus-leading-through-the-crisis"}]}}},{"mediaID":{"value":"CE"},"title":{"value":"Customer experience"},"description":{"value":"More insight into creating competitive advantage by putting customers first and managing their journeys."},"image":{"src":"/~/media/mckinsey/industries/public and social sector/our insights/customer experience/customer-experience_22934010_1536x864.jpg","alt":"Customer experience"},"moreLinks":{"jsonValue":{"value":[{"CssClass":"","LinkType":"internal","Target":"","Text":"Learn more","Title":"","Url":""}]}}},{"mediaID":{"value":"driversofdisruption"},"title":{"value":"Want to subscribe to \u003ca href=\"/features/mckinsey-center-for-future-mobility/our-insights/drivers-of-disruption\"\u003e\u003cem\u003eDrivers of Disruption\u003c/em\u003e\u003c/a\u003e?"},"description":{"value":""},"image":{"src":null,"alt":""},"moreLinks":{"jsonValue":{"value":[{"CssClass":"btn btn-fill","LinkType":"external","Target":"","Text":"Apple Podcasts","Title":"","Url":"https://podcasts.apple.com/us/podcast/drivers-of-disruption/id1687257324"},{"CssClass":"btn btn-fill","LinkType":"external","Target":"","Text":"Spotify","Title":"","Url":"https://open.spotify.com/show/5fWXzWmVKacc5I7u9CQsah?si=KQ-wUOSyQlqcnfexExiH9Q\u0026nd=1"}]}}},{"mediaID":{"value":"JOL"},"title":{"value":"The Journey of Leadership"},"description":{"value":"\u003cstrong\u003eHow CEOs Learn to Lead from the Inside Out\u003c/strong\u003e\u003cbr /\u003e\nBy \u003ca href=\"/our-people/dana-maor\"\u003eDana Maor\u003c/a\u003e, \u003ca href=\"https://www.linkedin.com/in/hanswernerkaas\"\u003eHans-Werner Kaas\u003c/a\u003e, \u003ca href=\"/our-people/kurt-strovink\"\u003eKurt Strovink\u003c/a\u003e, and \u003ca href=\"/our-people/ramesh-srinivasan\"\u003eRamesh Srinivasan\u003c/a\u003e"},"image":{"src":"/~/media/mckinsey/featured insights/mckinsey books/journey of leadership/journey-of-leadership-book-cover-1536x1536.jpg","alt":"The Journey of Leadership book cover"},"moreLinks":{"jsonValue":{"value":[{"CssClass":"","LinkType":"internal","Target":"","Text":"Learn more","Title":"","Url":"/featured-insights/mckinsey-on-books/journey-of-leadership"}]}}},{"mediaID":{"value":"surveys"},"title":{"value":"McKinsey\u0026rsquo;s original survey research"},"description":{"value":""},"image":{"src":"/~/media/mckinsey/featured insights/mckinsey global surveys/mck-global-survey-landing-1284442931-thumb-1536x1536.jpg","alt":"McKinsey Global Surveys"},"moreLinks":{"jsonValue":{"value":[{"CssClass":"","LinkType":"internal","Target":"","Text":"Read more","Title":"","Url":"/featured-insights/mckinsey-global-surveys"}]}}},{"mediaID":{"value":"healthcarepodcast"},"title":{"value":"Want to subscribe to \u003cem\u003eMcKinsey on Healthcare\u003c/em\u003e?"},"description":{"value":""},"image":{"src":null,"alt":""},"moreLinks":{"jsonValue":{"value":[{"CssClass":"btn btn-fill","LinkType":"external","Target":"","Text":"Apple Podcasts","Title":"","Url":"https://podcasts.apple.com/us/podcast/mckinsey-on-healthcare/id1469979748"},{"CssClass":"btn btn-fill","LinkType":"external","Target":"","Text":"Spotify","Title":"","Url":"https://open.spotify.com/show/3y1AgAilV4E5Jz0x4wD9O2?si=mLCTUQ2ORTS_LC5gW_htlA\u0026dl_branch=1"}]}}},{"mediaID":{"value":"myleadershipjourney"},"title":{"value":"My Leadership Journey"},"description":{"value":"A series of conversations with leaders about the people and experiences that shaped them"},"image":{"src":"/~/media/mckinsey/business functions/people and organizational performance/our insights/my leadership journey/my leadership journey collection page_131575496_thumb_1536x1536.jpg","alt":"My Leadership Journey"},"moreLinks":{"jsonValue":{"value":[{"CssClass":"","LinkType":"internal","Target":"","Text":"Learn more","Title":"","Url":"/capabilities/people-and-organizational-performance/our-insights/my-leadership-journey"}]}}},{"mediaID":{"value":"lessonsfromleaders"},"title":{"value":"State of Organizations: Lessons from leaders"},"description":{"value":""},"image":{"src":"/~/media/mckinsey/business functions/people and organizational performance/our insights/state of org lessons from leaders/soo-chapter-2-872507320-standard-1536x1536.jpg","alt":"\"\""},"moreLinks":{"jsonValue":{"value":[{"CssClass":"","LinkType":"internal","Target":"","Text":"Explore more interviews","Title":"","Url":"/capabilities/people-and-organizational-performance/our-insights/lessons-from-leaders"}]}}},{"mediaID":{"value":"stateoforg"},"title":{"value":"The State of Organizations 2023"},"description":{"value":""},"image":{"src":null,"alt":""},"moreLinks":{"jsonValue":{"value":[{"CssClass":"","LinkType":"internal","Target":"","Text":"Report overview","Title":"","Url":"/capabilities/people-and-organizational-performance/our-insights/the-state-of-organizations-2023"}]}}},{"mediaID":{"value":"titaniumeconomy"},"title":{"value":"The Titanium Economy"},"description":{"value":"How Industrial Technology Can Create a Better, Faster, Stronger America\u003cbr /\u003e\nBy \u003ca href=\"/our-people/asutosh-padhi\"\u003eAsutosh Padhi\u003c/a\u003e, Gaurav Batra, and \u003ca href=\"https://fernweh.com/nick-santhanam/\"\u003eNick Santhanam\u003c/a\u003e"},"image":{"src":"/~/media/mckinsey/featured insights/mckinsey books/the titanium economy/titanium_3d-1536x1536-hero-trans.png","alt":"The Titanium Economy Book cover"},"moreLinks":{"jsonValue":{"value":[{"CssClass":"","LinkType":"internal","Target":"","Text":"Learn more","Title":"","Url":"/featured-insights/mckinsey-on-books/the-titanium-economy"}]}}},{"mediaID":{"value":"GII"},"title":{"value":"Global Infrastructure Initiative"},"description":{"value":"Convening global leaders in infrastructure and capital projects in pursuit of new solutions"},"image":{"src":"/~/media/mckinsey/business functions/operations/our insights/the global infrastructure initiative/jpg-cpi-improving-construction-productivity-1536x864.jpg","alt":"Improving construction productivity"},"moreLinks":{"jsonValue":{"value":[{"CssClass":"","LinkType":"external","Target":"","Text":"Learn more","Title":"","Url":"http://www.globalinfrastructureinitiative.com/"}]}}},{"mediaID":{"value":"FOApodcast"},"title":{"value":"Subscribe to \u003ca href=\"/featured-insights/future-of-asia/future-of-asia-podcasts\"\u003e\u003cem\u003eFuture of Asia Podcasts\u003c/em\u003e\u003c/a\u003e"},"description":{"value":""},"image":{"src":null,"alt":""},"moreLinks":{"jsonValue":{"value":[{"CssClass":"btn btn-fill","LinkType":"external","Target":"","Text":"Apple Podcasts","Title":"","Url":"https://podcasts.apple.com/us/podcast/future-of-asia/id1480316959"},{"CssClass":"btn btn-fill","LinkType":"external","Target":"","Text":"Spotify","Title":"","Url":"https://open.spotify.com/show/2CZX4AR2DT4hQCFKJG0cCv?si=B20tkrOmQYymuCUhKCQOwA"}]}}},{"mediaID":{"value":"logisticsdisruptors"},"title":{"value":"Logistics Disruptors"},"description":{"value":"Meet the companies shaping how goods will move tomorrow."},"image":{"src":"/~/media/mckinsey/industries/travel logistics and infrastructure/our insights/logistics disruptors/tli-logistics-disruptors-1536x1536.jpeg","alt":"illustration autonomous truck driving off into the future"},"moreLinks":{"jsonValue":{"value":[{"CssClass":"","LinkType":"internal","Target":"","Text":"Learn more","Title":"","Url":"/industries/travel-logistics-and-infrastructure/our-insights/logistics-disruptors"}]}}},{"mediaID":{"value":"mckinseyexplainers"},"title":{"value":"Looking for direct answers to other complex questions?"},"description":{"value":""},"image":{"src":"/~/media/mckinsey/featured insights/mckinsey explainers/mckinseyexplainers-flat-thumb-1536x1536.jpg","alt":"Circular, white maze filled with white semicircles."},"moreLinks":{"jsonValue":{"value":[{"CssClass":"","LinkType":"internal","Target":"","Text":"Explore the full McKinsey Explainers series","Title":"","Url":"/featured-insights/mckinsey-explainers"}]}}},{"mediaID":{"value":"livesandlegacies"},"title":{"value":"\u003cem\u003e\u003ca href=\"/featured-insights/mckinsey-on-lives-and-legacies\"\u003eMcKinsey on Lives \u0026amp; Legacies\u003c/a\u003e\u003c/em\u003e"},"description":{"value":"Highlighting the lasting impact of leaders and executives"},"image":{"src":"/~/media/mckinsey/featured insights/mckinsey on lives and legacies/lives-and-legacies-landing-page-thumb-1536x1536.jpg","alt":"McKinsey on Lives \u0026 Legacies"},"moreLinks":{"jsonValue":{"value":[{"CssClass":"","LinkType":"internal","Target":"","Text":"Explore previous obituaries","Title":"","Url":"/featured-insights/mckinsey-on-lives-and-legacies"}]}}},{"mediaID":{"value":"Strategy"},"title":{"value":"Strategy in a digital age"},"description":{"value":"Our series on developing corporate and business-unit strategies in a digitally disrupted world."},"image":{"src":"/~/media/mckinsey/business functions/mckinsey digital/our insights/strategy in a digital age/digital-strategy_1536x864_flexpromoimage.jpg","alt":""},"moreLinks":{"jsonValue":{"value":[{"CssClass":"","LinkType":"internal","Target":"","Text":"View the collection","Title":"","Url":"/featured-insights/strategy-in-a-digital-age"}]}}},{"mediaID":{"value":"great"},"title":{"value":"The great re-make: Manufacturing for modern times"},"description":{"value":"This 21-article compendium gives practical insights for manufacturing leaders looking to keep a step ahead of today\u0026rsquo;s disruptions."},"image":{"src":"/~/media/mckinsey/business functions/operations/our insights/the great remake manufacturing for modern times/jpg-ops-dm-compendium-cover-1536x864.jpg","alt":"The great re-make: Manufacturing for modern times"},"moreLinks":{"jsonValue":{"value":[{"CssClass":"","LinkType":"internal","Target":"","Text":"View the collection","Title":"","Url":"/capabilities/operations/our-insights/the-great-re-make-manufacturing-for-modern-times-compendium"}]}}},{"mediaID":{"value":"futureofamerica"},"title":{"value":"Subscribe to the \u003cem\u003eFuture of America\u003c/em\u003e podcast"},"description":{"value":""},"image":{"src":null,"alt":""},"moreLinks":{"jsonValue":{"value":[{"CssClass":"btn btn-fill","LinkType":"external","Target":"_blank","Text":"Apple Podcasts","Title":"","Url":"https://podcasts.apple.com/us/podcast/future-of-america/id1616517376"},{"CssClass":"btn btn-fill","LinkType":"external","Target":"","Text":"Spotify","Title":"","Url":"https://open.spotify.com/show/0FpeyjojJTKf2hVtQ3otah"}]}}},{"mediaID":{"value":"governmentpodcast"},"title":{"value":"Subscribe to the \u003cem\u003eMcKinsey on Government\u003c/em\u003e podcast"},"description":{"value":""},"image":{"src":null,"alt":""},"moreLinks":{"jsonValue":{"value":[{"CssClass":"btn btn-fill","LinkType":"external","Target":"_blank","Text":"Apple Podcasts","Title":"","Url":"https://podcasts.apple.com/us/podcast/mckinsey-on-government/id1573645359"},{"CssClass":"btn btn-fill","LinkType":"external","Target":"","Text":"Spotify","Title":"","Url":"https://open.spotify.com/show/61XsO4B2ersroMft8t3KSq"}]}}},{"mediaID":{"value":"podcast"},"title":{"value":"Want to subscribe to the McKinsey Podcast?\n"},"description":{"value":"\u003cdiv class=\"cta-container -align-horizontal\"\u003e\n \u003ca href=\"#\" class=\"btn btn-fill\"\u003eGoogle Play\u003c/a\u003e\n \u003ca href=\"#\" class=\"btn btn-fill\"\u003eiTunes\u003c/a\u003e\n \u003ca href=\"#\" class=\"btn btn-fill\"\u003eStitcher\u003c/a\u003e\n \u003ca href=\"#\" class=\"btn mck-rss-icon\"\u003eRSS\u003c/a\u003e\n\u003c/div\u003e"},"image":{"src":null,"alt":""},"moreLinks":{"jsonValue":{"value":[]}}},{"mediaID":{"value":"operationspodcast"},"title":{"value":"Subscribe to the \u003cem\u003eMcKinsey Talks Operations\u003c/em\u003e podcast"},"description":{"value":""},"image":{"src":null,"alt":""},"moreLinks":{"jsonValue":{"value":[{"CssClass":"btn btn-fill","LinkType":"external","Target":"_blank","Text":"Apple Podcasts","Title":"","Url":"https://podcasts.apple.com/us/podcast/mckinsey-talks-operations/id1598128813"},{"CssClass":"btn btn-fill","LinkType":"external","Target":"","Text":"Spotify","Title":"","Url":"https://open.spotify.com/show/7LYQS2fxAoBmC3LgFNJbi8"}]}}},{"mediaID":{"value":"energyinsights"},"title":{"value":"Speak to an expert about our upstream capabilities"},"description":{"value":""},"image":{"src":"/~/media/mckinsey/industries/oil and gas/how we help clients/energy insights/newsletter-signup-promo-bar-1536x768.jpg","alt":"EI newsletter sign up"},"moreLinks":{"jsonValue":{"value":[{"CssClass":"btn btn-fill","LinkType":"internal","Target":"","Text":"Contact us","Title":"","Url":"/industries/oil-and-gas/how-we-help-clients/energy-solutions/contact-us"}]}}},{"mediaID":{"value":"buildingproductspodcast"},"title":{"value":"Subscribe to the \u003cem\u003eMcKinsey on Building Products\u003c/em\u003e podcast"},"description":{"value":""},"image":{"src":null,"alt":""},"moreLinks":{"jsonValue":{"value":[{"CssClass":"btn btn-fill","LinkType":"external","Target":"_blank","Text":"Apple Podcasts","Title":"","Url":"https://podcasts.apple.com/us/podcast/mckinsey-on-building-product/id1735607317"},{"CssClass":"btn btn-fill","LinkType":"external","Target":"_blank","Text":"Spotify","Title":"","Url":"https://open.spotify.com/show/79pncGJtBtzz7pFoR6OiH8"},{"CssClass":"","LinkType":"internal","Target":"","Text":"Listen to previous episodes","Title":"","Url":"/capabilities/people-and-organizational-performance/how-we-help-clients/mckinsey-academy/product-academy/mckinsey-on-building-product"}]}}},{"mediaID":{"value":"talentpodcast"},"title":{"value":"Subscribe to the \u003cem\u003eMcKinsey Talks Talent\u003c/em\u003e podcast"},"description":{"value":""},"image":{"src":null,"alt":""},"moreLinks":{"jsonValue":{"value":[{"CssClass":"btn btn-fill","LinkType":"external","Target":"_blank","Text":"Apple Podcasts","Title":"","Url":"https://podcasts.apple.com/us/podcast/mckinsey-talks-talent/id1491112396"},{"CssClass":"btn btn-fill","LinkType":"external","Target":"","Text":"Spotify","Title":"","Url":"https://open.spotify.com/show/5sW4eHEFMbOvAFJjMfN9a6"}]}}},{"mediaID":{"value":"outlook"},"title":{"value":"A longer, more comprehensive version of our Global Oil Supply \u0026 Demand Outlook is available for purchase upon request.\n"},"description":{"value":""},"image":{"src":null,"alt":""},"moreLinks":{"jsonValue":{"value":[{"CssClass":"btn btn-fill","LinkType":"internal","Target":"","Text":"Contact us","Title":"","Url":"/industries/oil-and-gas/how-we-help-clients/energy-solutions/contact-us"}]}}},{"mediaID":{"value":"talentpodcast2"},"title":{"value":"McKinsey Talks Talent Podcast"},"description":{"value":"Bryan Hancock, Brooke Weddle, and other talent experts help you navigate a fast-changing landscape and prepare for the future of work by making talent a competitive advantage."},"image":{"src":"/~/media/mckinsey/business functions/people and organizational performance/mckinsey talks talent/mck-mtt-disruptor-full-1536x864-smoke-may-2021.jpg","alt":"McKinsey Talks Talent"},"moreLinks":{"jsonValue":{"value":[{"CssClass":"","LinkType":"internal","Target":"","Text":"See more episodes","Title":"","Url":"/capabilities/people-and-organizational-performance/mckinsey-talks-talent-podcast"}]}}},{"mediaID":{"value":"investing"},"title":{"value":"Subscribe to McKinsey on Investing"},"description":{"value":""},"image":{"src":null,"alt":""},"moreLinks":{"jsonValue":{"value":[{"CssClass":"btn btn-fill","LinkType":"internal","Target":"","Text":"Learn more","Title":"","Url":"/industries/private-capital/our-insights/mckinsey-on-investing-issue-1/contact"}]}}},{"mediaID":{"value":"women matter"},"title":{"value":"Subscribe to McKinsey \u003cem\u003eWomen Matter Canada\u003c/em\u003e"},"description":{"value":"Please subscribe if your organization is interested in learning about participating in future research."},"image":{"src":null,"alt":""},"moreLinks":{"jsonValue":{"value":[{"CssClass":"btn btn-fill","LinkType":"internal","Target":"","Text":"Learn more","Title":"","Url":""}]}}},{"mediaID":{"value":"customs"},"title":{"value":"This article was written as part of a larger research effort to identify opportunities for improvement and innovation at customs agencies globally."},"description":{"value":""},"image":{"src":null,"alt":""},"moreLinks":{"jsonValue":{"value":[{"CssClass":"btn btn-fill","LinkType":"internal","Target":"","Text":"Learn more","Title":"","Url":"/industries/public-sector/our-insights/contact"}]}}},{"mediaID":{"value":"tech"},"title":{"value":"\u003cspan class=\"disrupt-ab\"\u003eCreating value beyond the hype\u003c/span\u003e"},"description":{"value":"Let\u0026rsquo;s deliver on the promise of technology from strategy to scale."},"image":{"src":"/~/media/mckinsey/business functions/mckinsey digital/how we help clients/2024/never-just-tech_thumb_1536x1536.jpg","alt":"Never just tech"},"moreLinks":{"jsonValue":{"value":[{"CssClass":"","LinkType":"external","Target":"","Text":"Get started","Title":"","Url":"https://www.mckinsey.com/capabilities/mckinsey-digital/how-we-help-clients?cid=njt-ste-crk-mbm-mbm--2403-promo01-njt-bam-web"}]}}},{"mediaID":{"value":"sust"},"title":{"value":"\u003cspan class=\"disrupt-ab\"\u003eMove from plans to progress.\u003c/span\u003e"},"description":{"value":"Sustainability matters. Together we’ll make it real."},"image":{"src":"/~/media/mckinsey/business functions/sustainability/how we help clients/sustainability-campaign_promo-thumb_1536x1536.jpg","alt":"Move from plans to progress."},"moreLinks":{"jsonValue":{"value":[{"CssClass":"","LinkType":"external","Target":"","Text":"Get started","Title":"","Url":"https://www.mckinsey.com/capabilities/sustainability/how-we-help-clients/?cid=susti24-ste-crk-mbm-m01-other-glb-web"}]}}},{"mediaID":{"value":"nwowpodcast"},"title":{"value":"Subscribe to the New World of Work podcast"},"description":{"value":""},"image":{"src":null,"alt":""},"moreLinks":{"jsonValue":{"value":[{"CssClass":"btn btn-fill","LinkType":"external","Target":"_blank","Text":"Apple Podcasts","Title":"","Url":"https://itunes.apple.com/us/podcast/the-new-world-of-work/id1319502839?mt=2"},{"CssClass":"btn btn-fill","LinkType":"external","Target":"","Text":"Spotify","Title":"","Url":"https://open.spotify.com/show/3S6fHLW6mcbDJDp1wkArKI"}]}}},{"mediaID":{"value":"globalpodcast"},"title":{"value":"Want to subscribe to \u003cem\u003eThe McKinsey Podcast\u003c/em\u003e?"},"description":{"value":""},"image":{"src":null,"alt":""},"moreLinks":{"jsonValue":{"value":[{"CssClass":"btn btn-fill","LinkType":"external","Target":"_blank","Text":"Apple Podcasts","Title":"","Url":"https://itunes.apple.com/us/podcast/the-mckinsey-podcast/id285260960?mt=2"},{"CssClass":"btn btn-fill","LinkType":"external","Target":"","Text":"Spotify","Title":"","Url":"https://open.spotify.com/show/4dyjRTP9xzEUPgleo6XjXI"},{"CssClass":"btn btn-fill","LinkType":"external","Target":"_blank","Text":"YouTube","Title":"","Url":"https://www.youtube.com/@McKinsey/podcasts"},{"CssClass":"","LinkType":"internal","Target":"","Text":"Listen to previous episodes","Title":"","Url":"/featured-insights/mckinsey-podcast"}]}}},{"mediaID":{"value":"Qaudio"},"title":{"value":"Discover and subscribe to McKinsey Quarterly Audio"},"description":{"value":"\u003cstrong\u003eFive ways to subscribe:\u003c/strong\u003e"},"image":{"src":null,"alt":""},"moreLinks":{"jsonValue":{"value":[{"CssClass":"btn btn-fill","LinkType":"external","Target":"_blank","Text":"Apple Podcasts","Title":"","Url":"https://itunes.apple.com/us/podcast/mckinsey-quarterly-audio/id1437925222?mt=2"},{"CssClass":"btn btn-fill","LinkType":"external","Target":"","Text":"Spotify","Title":"","Url":"https://open.spotify.com/show/5mb2zYJnrPcRrML8gx5doD"}]}}},{"mediaID":{"value":"shortlist"},"title":{"value":"Subscribe to the Shortlist"},"description":{"value":"McKinsey\u0026rsquo;s new weekly newsletter, featuring must-read content on a range of topics, every Friday"},"image":{"src":"/~/media/mckinsey/email/shortlist/template/shortlist_promo-interrupter-849120982_1536x1536.jpg","alt":"the Shortlist"},"moreLinks":{"jsonValue":{"value":[{"CssClass":"","LinkType":"internal","Target":"","Text":"Get the Shortlist in your inbox","Title":"","Url":"/user-registration/manage-account/edit-subscriptions"}]}}},{"mediaID":{"value":"retailpodcast"},"title":{"value":"Subscribe to the \u003cem\u003eMcKinsey on Consumer and Retail\u003c/em\u003e podcast"},"description":{"value":""},"image":{"src":null,"alt":""},"moreLinks":{"jsonValue":{"value":[{"CssClass":"btn btn-fill","LinkType":"external","Target":"_blank","Text":"Apple Podcasts","Title":"","Url":"https://podcasts.apple.com/us/podcast/mckinsey-on-consumer-and-retail/id1526250428"},{"CssClass":"btn btn-fill","LinkType":"external","Target":"_blank","Text":"Spotify","Title":"","Url":"https://open.spotify.com/show/41D4bro8PPDOGhEgOEJyHB?si=7l3SeYCZQD-7v3toqd9fHQ"}]}}},{"mediaID":{"value":"insurancepodcast"},"title":{"value":"Want to subscribe to the \u003cem\u003eMcKinsey on Insurance\u003c/em\u003e podcast?"},"description":{"value":""},"image":{"src":null,"alt":""},"moreLinks":{"jsonValue":{"value":[{"CssClass":"btn btn-fill","LinkType":"external","Target":"_blank","Text":"Apple Podcasts","Title":"","Url":"https://podcasts.apple.com/us/podcast/mckinsey-on-insurance/id1533196359"},{"CssClass":"btn btn-fill","LinkType":"external","Target":"_blank","Text":"Spotify","Title":"","Url":"https://open.spotify.com/show/5gZ6YKF7n7C9zWaER4mFZ5"}]}}},{"mediaID":{"value":"strategypodcast"},"title":{"value":"Subscribe to the \u003cem\u003eInside the Strategy Room\u003c/em\u003e podcast"},"description":{"value":""},"image":{"src":null,"alt":""},"moreLinks":{"jsonValue":{"value":[{"CssClass":"btn btn-fill","LinkType":"external","Target":"_blank","Text":"Apple Podcasts","Title":"","Url":"https://itunes.apple.com/us/podcast/inside-the-strategy-room/id1422814215?mt=2"},{"CssClass":"btn btn-fill","LinkType":"external","Target":"","Text":"Spotify","Title":"","Url":"https://open.spotify.com/show/4TcFfiQ0e6OYuc5kRDLxqj"}]}}},{"mediaID":{"value":"mgipodcast"},"title":{"value":"Subscribe to the \u003cem\u003eForward Thinking\u003c/em\u003e podcast"},"description":{"value":""},"image":{"src":null,"alt":""},"moreLinks":{"jsonValue":{"value":[{"CssClass":"btn btn-fill","LinkType":"external","Target":"_blank","Text":"Apple Podcasts","Title":"","Url":"https://podcasts.apple.com/us/podcast/forward-thinking/id1319502839"},{"CssClass":"btn btn-fill","LinkType":"external","Target":"_blank","Text":"Spotify","Title":"","Url":"https://open.spotify.com/show/3S6fHLW6mcbDJDp1wkArKI?si=0zvPmiCsQ7SBAPanfpY4aA\u0026nd=1"},{"CssClass":"","LinkType":"internal","Target":"","Text":"Listen to previous episodes","Title":"","Url":"/mgi/forward-thinking"}]}}}]},"globalDisruptors":{"disruptor1up":{"results":[{"mediaID":{"value":"cop"},"title":{"value":"Explore COP29 with McKinsey"},"description":{"value":"Join our series of dynamic virtual events during COP29. Discover new research, practical strategies, and collaborations across sectors to move climate action beyond why to how."},"image":{"src":"/~/media/mckinsey/business functions/sustainability/how we help clients/cop/cop29/cop_29_thumb.jpg","alt":"McKinsey at COP29"},"moreLinks":{"jsonValue":{"value":[{"CssClass":"","LinkType":"external","Target":"","Text":"Register","Title":"","Url":"https://mckinsey.cventevents.com/Q2MVE1?rt=RrQJ1iBSfEiqSK7N9VAGCw\u0026RefId=cop"}]}}},{"mediaID":{"value":"attheedge"},"title":{"value":"Subscribe to the \u003cem\u003eAt the Edge\u003c/em\u003e podcast"},"description":{"value":""},"image":{"src":null,"alt":""},"moreLinks":{"jsonValue":{"value":[{"CssClass":"btn btn-fill","LinkType":"external","Target":"","Text":"Apple Podcasts","Title":"","Url":"https://podcasts.apple.com/us/podcast/at-the-edge/id1612870236"},{"CssClass":"btn btn-fill","LinkType":"external","Target":"","Text":"Spotify","Title":"","Url":"https://open.spotify.com/show/3SnUcezwM20nKnzqvUl6iF"}]}}},{"mediaID":{"value":"authortalks"},"title":{"value":"Visit \u003cem\u003e\u003ca href=\"/featured-insights/mckinsey-on-books/author-talks\"\u003eAuthor Talks\u003c/a\u003e\u003c/em\u003e to see the full series."},"description":{"value":""},"image":{"src":"/~/media/mckinsey/featured insights/mckinsey books/author talks/authortalks_1536x1536_v1.png","alt":"Author Talks"},"moreLinks":{"jsonValue":{"value":[{"CssClass":"","LinkType":"internal","Target":"","Text":"Learn more","Title":"","Url":"/featured-insights/mckinsey-on-books/author-talks"}]}}},{"mediaID":{"value":"COVID"},"title":{"value":"Explore McKinsey’s ongoing coverage of the pandemic"},"description":{"value":""},"image":{"src":"/~/media/mckinsey/business functions/risk/our insights/covid 19 implications for business/implications-briefing-note-4-standard-1536x1536.jpg","alt":"COVID-19: Implications for business"},"moreLinks":{"jsonValue":{"value":[{"CssClass":"","LinkType":"internal","Target":"","Text":"Read more about the business impact of coronavirus and how organizations can respond","Title":"","Url":"/featured-insights/coronavirus-leading-through-the-crisis"}]}}},{"mediaID":{"value":"CE"},"title":{"value":"Customer experience"},"description":{"value":"More insight into creating competitive advantage by putting customers first and managing their journeys."},"image":{"src":"/~/media/mckinsey/industries/public and social sector/our insights/customer experience/customer-experience_22934010_1536x864.jpg","alt":"Customer experience"},"moreLinks":{"jsonValue":{"value":[{"CssClass":"","LinkType":"internal","Target":"","Text":"Learn more","Title":"","Url":""}]}}},{"mediaID":{"value":"driversofdisruption"},"title":{"value":"Want to subscribe to \u003ca href=\"/features/mckinsey-center-for-future-mobility/our-insights/drivers-of-disruption\"\u003e\u003cem\u003eDrivers of Disruption\u003c/em\u003e\u003c/a\u003e?"},"description":{"value":""},"image":{"src":null,"alt":""},"moreLinks":{"jsonValue":{"value":[{"CssClass":"btn btn-fill","LinkType":"external","Target":"","Text":"Apple Podcasts","Title":"","Url":"https://podcasts.apple.com/us/podcast/drivers-of-disruption/id1687257324"},{"CssClass":"btn btn-fill","LinkType":"external","Target":"","Text":"Spotify","Title":"","Url":"https://open.spotify.com/show/5fWXzWmVKacc5I7u9CQsah?si=KQ-wUOSyQlqcnfexExiH9Q\u0026nd=1"}]}}},{"mediaID":{"value":"JOL"},"title":{"value":"The Journey of Leadership"},"description":{"value":"\u003cstrong\u003eHow CEOs Learn to Lead from the Inside Out\u003c/strong\u003e\u003cbr /\u003e\nBy \u003ca href=\"/our-people/dana-maor\"\u003eDana Maor\u003c/a\u003e, \u003ca href=\"https://www.linkedin.com/in/hanswernerkaas\"\u003eHans-Werner Kaas\u003c/a\u003e, \u003ca href=\"/our-people/kurt-strovink\"\u003eKurt Strovink\u003c/a\u003e, and \u003ca href=\"/our-people/ramesh-srinivasan\"\u003eRamesh Srinivasan\u003c/a\u003e"},"image":{"src":"/~/media/mckinsey/featured insights/mckinsey books/journey of leadership/journey-of-leadership-book-cover-1536x1536.jpg","alt":"The Journey of Leadership book cover"},"moreLinks":{"jsonValue":{"value":[{"CssClass":"","LinkType":"internal","Target":"","Text":"Learn more","Title":"","Url":"/featured-insights/mckinsey-on-books/journey-of-leadership"}]}}},{"mediaID":{"value":"surveys"},"title":{"value":"McKinsey\u0026rsquo;s original survey research"},"description":{"value":""},"image":{"src":"/~/media/mckinsey/featured insights/mckinsey global surveys/mck-global-survey-landing-1284442931-thumb-1536x1536.jpg","alt":"McKinsey Global Surveys"},"moreLinks":{"jsonValue":{"value":[{"CssClass":"","LinkType":"internal","Target":"","Text":"Read more","Title":"","Url":"/featured-insights/mckinsey-global-surveys"}]}}},{"mediaID":{"value":"healthcarepodcast"},"title":{"value":"Want to subscribe to \u003cem\u003eMcKinsey on Healthcare\u003c/em\u003e?"},"description":{"value":""},"image":{"src":null,"alt":""},"moreLinks":{"jsonValue":{"value":[{"CssClass":"btn btn-fill","LinkType":"external","Target":"","Text":"Apple Podcasts","Title":"","Url":"https://podcasts.apple.com/us/podcast/mckinsey-on-healthcare/id1469979748"},{"CssClass":"btn btn-fill","LinkType":"external","Target":"","Text":"Spotify","Title":"","Url":"https://open.spotify.com/show/3y1AgAilV4E5Jz0x4wD9O2?si=mLCTUQ2ORTS_LC5gW_htlA\u0026dl_branch=1"}]}}},{"mediaID":{"value":"myleadershipjourney"},"title":{"value":"My Leadership Journey"},"description":{"value":"A series of conversations with leaders about the people and experiences that shaped them"},"image":{"src":"/~/media/mckinsey/business functions/people and organizational performance/our insights/my leadership journey/my leadership journey collection page_131575496_thumb_1536x1536.jpg","alt":"My Leadership Journey"},"moreLinks":{"jsonValue":{"value":[{"CssClass":"","LinkType":"internal","Target":"","Text":"Learn more","Title":"","Url":"/capabilities/people-and-organizational-performance/our-insights/my-leadership-journey"}]}}},{"mediaID":{"value":"lessonsfromleaders"},"title":{"value":"State of Organizations: Lessons from leaders"},"description":{"value":""},"image":{"src":"/~/media/mckinsey/business functions/people and organizational performance/our insights/state of org lessons from leaders/soo-chapter-2-872507320-standard-1536x1536.jpg","alt":"\"\""},"moreLinks":{"jsonValue":{"value":[{"CssClass":"","LinkType":"internal","Target":"","Text":"Explore more interviews","Title":"","Url":"/capabilities/people-and-organizational-performance/our-insights/lessons-from-leaders"}]}}},{"mediaID":{"value":"stateoforg"},"title":{"value":"The State of Organizations 2023"},"description":{"value":""},"image":{"src":null,"alt":""},"moreLinks":{"jsonValue":{"value":[{"CssClass":"","LinkType":"internal","Target":"","Text":"Report overview","Title":"","Url":"/capabilities/people-and-organizational-performance/our-insights/the-state-of-organizations-2023"}]}}},{"mediaID":{"value":"titaniumeconomy"},"title":{"value":"The Titanium Economy"},"description":{"value":"How Industrial Technology Can Create a Better, Faster, Stronger America\u003cbr /\u003e\nBy \u003ca href=\"/our-people/asutosh-padhi\"\u003eAsutosh Padhi\u003c/a\u003e, Gaurav Batra, and \u003ca href=\"https://fernweh.com/nick-santhanam/\"\u003eNick Santhanam\u003c/a\u003e"},"image":{"src":"/~/media/mckinsey/featured insights/mckinsey books/the titanium economy/titanium_3d-1536x1536-hero-trans.png","alt":"The Titanium Economy Book cover"},"moreLinks":{"jsonValue":{"value":[{"CssClass":"","LinkType":"internal","Target":"","Text":"Learn more","Title":"","Url":"/featured-insights/mckinsey-on-books/the-titanium-economy"}]}}},{"mediaID":{"value":"GII"},"title":{"value":"Global Infrastructure Initiative"},"description":{"value":"Convening global leaders in infrastructure and capital projects in pursuit of new solutions"},"image":{"src":"/~/media/mckinsey/business functions/operations/our insights/the global infrastructure initiative/jpg-cpi-improving-construction-productivity-1536x864.jpg","alt":"Improving construction productivity"},"moreLinks":{"jsonValue":{"value":[{"CssClass":"","LinkType":"external","Target":"","Text":"Learn more","Title":"","Url":"http://www.globalinfrastructureinitiative.com/"}]}}},{"mediaID":{"value":"FOApodcast"},"title":{"value":"Subscribe to \u003ca href=\"/featured-insights/future-of-asia/future-of-asia-podcasts\"\u003e\u003cem\u003eFuture of Asia Podcasts\u003c/em\u003e\u003c/a\u003e"},"description":{"value":""},"image":{"src":null,"alt":""},"moreLinks":{"jsonValue":{"value":[{"CssClass":"btn btn-fill","LinkType":"external","Target":"","Text":"Apple Podcasts","Title":"","Url":"https://podcasts.apple.com/us/podcast/future-of-asia/id1480316959"},{"CssClass":"btn btn-fill","LinkType":"external","Target":"","Text":"Spotify","Title":"","Url":"https://open.spotify.com/show/2CZX4AR2DT4hQCFKJG0cCv?si=B20tkrOmQYymuCUhKCQOwA"}]}}},{"mediaID":{"value":"logisticsdisruptors"},"title":{"value":"Logistics Disruptors"},"description":{"value":"Meet the companies shaping how goods will move tomorrow."},"image":{"src":"/~/media/mckinsey/industries/travel logistics and infrastructure/our insights/logistics disruptors/tli-logistics-disruptors-1536x1536.jpeg","alt":"illustration autonomous truck driving off into the future"},"moreLinks":{"jsonValue":{"value":[{"CssClass":"","LinkType":"internal","Target":"","Text":"Learn more","Title":"","Url":"/industries/travel-logistics-and-infrastructure/our-insights/logistics-disruptors"}]}}},{"mediaID":{"value":"mckinseyexplainers"},"title":{"value":"Looking for direct answers to other complex questions?"},"description":{"value":""},"image":{"src":"/~/media/mckinsey/featured insights/mckinsey explainers/mckinseyexplainers-flat-thumb-1536x1536.jpg","alt":"Circular, white maze filled with white semicircles."},"moreLinks":{"jsonValue":{"value":[{"CssClass":"","LinkType":"internal","Target":"","Text":"Explore the full McKinsey Explainers series","Title":"","Url":"/featured-insights/mckinsey-explainers"}]}}},{"mediaID":{"value":"livesandlegacies"},"title":{"value":"\u003cem\u003e\u003ca href=\"/featured-insights/mckinsey-on-lives-and-legacies\"\u003eMcKinsey on Lives \u0026amp; Legacies\u003c/a\u003e\u003c/em\u003e"},"description":{"value":"Highlighting the lasting impact of leaders and executives"},"image":{"src":"/~/media/mckinsey/featured insights/mckinsey on lives and legacies/lives-and-legacies-landing-page-thumb-1536x1536.jpg","alt":"McKinsey on Lives \u0026 Legacies"},"moreLinks":{"jsonValue":{"value":[{"CssClass":"","LinkType":"internal","Target":"","Text":"Explore previous obituaries","Title":"","Url":"/featured-insights/mckinsey-on-lives-and-legacies"}]}}},{"mediaID":{"value":"Strategy"},"title":{"value":"Strategy in a digital age"},"description":{"value":"Our series on developing corporate and business-unit strategies in a digitally disrupted world."},"image":{"src":"/~/media/mckinsey/business functions/mckinsey digital/our insights/strategy in a digital age/digital-strategy_1536x864_flexpromoimage.jpg","alt":""},"moreLinks":{"jsonValue":{"value":[{"CssClass":"","LinkType":"internal","Target":"","Text":"View the collection","Title":"","Url":"/featured-insights/strategy-in-a-digital-age"}]}}},{"mediaID":{"value":"great"},"title":{"value":"The great re-make: Manufacturing for modern times"},"description":{"value":"This 21-article compendium gives practical insights for manufacturing leaders looking to keep a step ahead of today\u0026rsquo;s disruptions."},"image":{"src":"/~/media/mckinsey/business functions/operations/our insights/the great remake manufacturing for modern times/jpg-ops-dm-compendium-cover-1536x864.jpg","alt":"The great re-make: Manufacturing for modern times"},"moreLinks":{"jsonValue":{"value":[{"CssClass":"","LinkType":"internal","Target":"","Text":"View the collection","Title":"","Url":"/capabilities/operations/our-insights/the-great-re-make-manufacturing-for-modern-times-compendium"}]}}},{"mediaID":{"value":"futureofamerica"},"title":{"value":"Subscribe to the \u003cem\u003eFuture of America\u003c/em\u003e podcast"},"description":{"value":""},"image":{"src":null,"alt":""},"moreLinks":{"jsonValue":{"value":[{"CssClass":"btn btn-fill","LinkType":"external","Target":"_blank","Text":"Apple Podcasts","Title":"","Url":"https://podcasts.apple.com/us/podcast/future-of-america/id1616517376"},{"CssClass":"btn btn-fill","LinkType":"external","Target":"","Text":"Spotify","Title":"","Url":"https://open.spotify.com/show/0FpeyjojJTKf2hVtQ3otah"}]}}},{"mediaID":{"value":"governmentpodcast"},"title":{"value":"Subscribe to the \u003cem\u003eMcKinsey on Government\u003c/em\u003e podcast"},"description":{"value":""},"image":{"src":null,"alt":""},"moreLinks":{"jsonValue":{"value":[{"CssClass":"btn btn-fill","LinkType":"external","Target":"_blank","Text":"Apple Podcasts","Title":"","Url":"https://podcasts.apple.com/us/podcast/mckinsey-on-government/id1573645359"},{"CssClass":"btn btn-fill","LinkType":"external","Target":"","Text":"Spotify","Title":"","Url":"https://open.spotify.com/show/61XsO4B2ersroMft8t3KSq"}]}}},{"mediaID":{"value":"podcast"},"title":{"value":"Want to subscribe to the McKinsey Podcast?\n"},"description":{"value":"\u003cdiv class=\"cta-container -align-horizontal\"\u003e\n \u003ca href=\"#\" class=\"btn btn-fill\"\u003eGoogle Play\u003c/a\u003e\n \u003ca href=\"#\" class=\"btn btn-fill\"\u003eiTunes\u003c/a\u003e\n \u003ca href=\"#\" class=\"btn btn-fill\"\u003eStitcher\u003c/a\u003e\n \u003ca href=\"#\" class=\"btn mck-rss-icon\"\u003eRSS\u003c/a\u003e\n\u003c/div\u003e"},"image":{"src":null,"alt":""},"moreLinks":{"jsonValue":{"value":[]}}},{"mediaID":{"value":"operationspodcast"},"title":{"value":"Subscribe to the \u003cem\u003eMcKinsey Talks Operations\u003c/em\u003e podcast"},"description":{"value":""},"image":{"src":null,"alt":""},"moreLinks":{"jsonValue":{"value":[{"CssClass":"btn btn-fill","LinkType":"external","Target":"_blank","Text":"Apple Podcasts","Title":"","Url":"https://podcasts.apple.com/us/podcast/mckinsey-talks-operations/id1598128813"},{"CssClass":"btn btn-fill","LinkType":"external","Target":"","Text":"Spotify","Title":"","Url":"https://open.spotify.com/show/7LYQS2fxAoBmC3LgFNJbi8"}]}}},{"mediaID":{"value":"energyinsights"},"title":{"value":"Speak to an expert about our upstream capabilities"},"description":{"value":""},"image":{"src":"/~/media/mckinsey/industries/oil and gas/how we help clients/energy insights/newsletter-signup-promo-bar-1536x768.jpg","alt":"EI newsletter sign up"},"moreLinks":{"jsonValue":{"value":[{"CssClass":"btn btn-fill","LinkType":"internal","Target":"","Text":"Contact us","Title":"","Url":"/industries/oil-and-gas/how-we-help-clients/energy-solutions/contact-us"}]}}},{"mediaID":{"value":"buildingproductspodcast"},"title":{"value":"Subscribe to the \u003cem\u003eMcKinsey on Building Products\u003c/em\u003e podcast"},"description":{"value":""},"image":{"src":null,"alt":""},"moreLinks":{"jsonValue":{"value":[{"CssClass":"btn btn-fill","LinkType":"external","Target":"_blank","Text":"Apple Podcasts","Title":"","Url":"https://podcasts.apple.com/us/podcast/mckinsey-on-building-product/id1735607317"},{"CssClass":"btn btn-fill","LinkType":"external","Target":"_blank","Text":"Spotify","Title":"","Url":"https://open.spotify.com/show/79pncGJtBtzz7pFoR6OiH8"},{"CssClass":"","LinkType":"internal","Target":"","Text":"Listen to previous episodes","Title":"","Url":"/capabilities/people-and-organizational-performance/how-we-help-clients/mckinsey-academy/product-academy/mckinsey-on-building-product"}]}}},{"mediaID":{"value":"talentpodcast"},"title":{"value":"Subscribe to the \u003cem\u003eMcKinsey Talks Talent\u003c/em\u003e podcast"},"description":{"value":""},"image":{"src":null,"alt":""},"moreLinks":{"jsonValue":{"value":[{"CssClass":"btn btn-fill","LinkType":"external","Target":"_blank","Text":"Apple Podcasts","Title":"","Url":"https://podcasts.apple.com/us/podcast/mckinsey-talks-talent/id1491112396"},{"CssClass":"btn btn-fill","LinkType":"external","Target":"","Text":"Spotify","Title":"","Url":"https://open.spotify.com/show/5sW4eHEFMbOvAFJjMfN9a6"}]}}},{"mediaID":{"value":"outlook"},"title":{"value":"A longer, more comprehensive version of our Global Oil Supply \u0026 Demand Outlook is available for purchase upon request.\n"},"description":{"value":""},"image":{"src":null,"alt":""},"moreLinks":{"jsonValue":{"value":[{"CssClass":"btn btn-fill","LinkType":"internal","Target":"","Text":"Contact us","Title":"","Url":"/industries/oil-and-gas/how-we-help-clients/energy-solutions/contact-us"}]}}},{"mediaID":{"value":"talentpodcast2"},"title":{"value":"McKinsey Talks Talent Podcast"},"description":{"value":"Bryan Hancock, Brooke Weddle, and other talent experts help you navigate a fast-changing landscape and prepare for the future of work by making talent a competitive advantage."},"image":{"src":"/~/media/mckinsey/business functions/people and organizational performance/mckinsey talks talent/mck-mtt-disruptor-full-1536x864-smoke-may-2021.jpg","alt":"McKinsey Talks Talent"},"moreLinks":{"jsonValue":{"value":[{"CssClass":"","LinkType":"internal","Target":"","Text":"See more episodes","Title":"","Url":"/capabilities/people-and-organizational-performance/mckinsey-talks-talent-podcast"}]}}},{"mediaID":{"value":"investing"},"title":{"value":"Subscribe to McKinsey on Investing"},"description":{"value":""},"image":{"src":null,"alt":""},"moreLinks":{"jsonValue":{"value":[{"CssClass":"btn btn-fill","LinkType":"internal","Target":"","Text":"Learn more","Title":"","Url":"/industries/private-capital/our-insights/mckinsey-on-investing-issue-1/contact"}]}}},{"mediaID":{"value":"women matter"},"title":{"value":"Subscribe to McKinsey \u003cem\u003eWomen Matter Canada\u003c/em\u003e"},"description":{"value":"Please subscribe if your organization is interested in learning about participating in future research."},"image":{"src":null,"alt":""},"moreLinks":{"jsonValue":{"value":[{"CssClass":"btn btn-fill","LinkType":"internal","Target":"","Text":"Learn more","Title":"","Url":""}]}}},{"mediaID":{"value":"customs"},"title":{"value":"This article was written as part of a larger research effort to identify opportunities for improvement and innovation at customs agencies globally."},"description":{"value":""},"image":{"src":null,"alt":""},"moreLinks":{"jsonValue":{"value":[{"CssClass":"btn btn-fill","LinkType":"internal","Target":"","Text":"Learn more","Title":"","Url":"/industries/public-sector/our-insights/contact"}]}}},{"mediaID":{"value":"tech"},"title":{"value":"\u003cspan class=\"disrupt-ab\"\u003eCreating value beyond the hype\u003c/span\u003e"},"description":{"value":"Let\u0026rsquo;s deliver on the promise of technology from strategy to scale."},"image":{"src":"/~/media/mckinsey/business functions/mckinsey digital/how we help clients/2024/never-just-tech_thumb_1536x1536.jpg","alt":"Never just tech"},"moreLinks":{"jsonValue":{"value":[{"CssClass":"","LinkType":"external","Target":"","Text":"Get started","Title":"","Url":"https://www.mckinsey.com/capabilities/mckinsey-digital/how-we-help-clients?cid=njt-ste-crk-mbm-mbm--2403-promo01-njt-bam-web"}]}}},{"mediaID":{"value":"sust"},"title":{"value":"\u003cspan class=\"disrupt-ab\"\u003eMove from plans to progress.\u003c/span\u003e"},"description":{"value":"Sustainability matters. Together we’ll make it real."},"image":{"src":"/~/media/mckinsey/business functions/sustainability/how we help clients/sustainability-campaign_promo-thumb_1536x1536.jpg","alt":"Move from plans to progress."},"moreLinks":{"jsonValue":{"value":[{"CssClass":"","LinkType":"external","Target":"","Text":"Get started","Title":"","Url":"https://www.mckinsey.com/capabilities/sustainability/how-we-help-clients/?cid=susti24-ste-crk-mbm-m01-other-glb-web"}]}}},{"mediaID":{"value":"nwowpodcast"},"title":{"value":"Subscribe to the New World of Work podcast"},"description":{"value":""},"image":{"src":null,"alt":""},"moreLinks":{"jsonValue":{"value":[{"CssClass":"btn btn-fill","LinkType":"external","Target":"_blank","Text":"Apple Podcasts","Title":"","Url":"https://itunes.apple.com/us/podcast/the-new-world-of-work/id1319502839?mt=2"},{"CssClass":"btn btn-fill","LinkType":"external","Target":"","Text":"Spotify","Title":"","Url":"https://open.spotify.com/show/3S6fHLW6mcbDJDp1wkArKI"}]}}},{"mediaID":{"value":"globalpodcast"},"title":{"value":"Want to subscribe to \u003cem\u003eThe McKinsey Podcast\u003c/em\u003e?"},"description":{"value":""},"image":{"src":null,"alt":""},"moreLinks":{"jsonValue":{"value":[{"CssClass":"btn btn-fill","LinkType":"external","Target":"_blank","Text":"Apple Podcasts","Title":"","Url":"https://itunes.apple.com/us/podcast/the-mckinsey-podcast/id285260960?mt=2"},{"CssClass":"btn btn-fill","LinkType":"external","Target":"","Text":"Spotify","Title":"","Url":"https://open.spotify.com/show/4dyjRTP9xzEUPgleo6XjXI"},{"CssClass":"btn btn-fill","LinkType":"external","Target":"_blank","Text":"YouTube","Title":"","Url":"https://www.youtube.com/@McKinsey/podcasts"},{"CssClass":"","LinkType":"internal","Target":"","Text":"Listen to previous episodes","Title":"","Url":"/featured-insights/mckinsey-podcast"}]}}},{"mediaID":{"value":"Qaudio"},"title":{"value":"Discover and subscribe to McKinsey Quarterly Audio"},"description":{"value":"\u003cstrong\u003eFive ways to subscribe:\u003c/strong\u003e"},"image":{"src":null,"alt":""},"moreLinks":{"jsonValue":{"value":[{"CssClass":"btn btn-fill","LinkType":"external","Target":"_blank","Text":"Apple Podcasts","Title":"","Url":"https://itunes.apple.com/us/podcast/mckinsey-quarterly-audio/id1437925222?mt=2"},{"CssClass":"btn btn-fill","LinkType":"external","Target":"","Text":"Spotify","Title":"","Url":"https://open.spotify.com/show/5mb2zYJnrPcRrML8gx5doD"}]}}},{"mediaID":{"value":"shortlist"},"title":{"value":"Subscribe to the Shortlist"},"description":{"value":"McKinsey\u0026rsquo;s new weekly newsletter, featuring must-read content on a range of topics, every Friday"},"image":{"src":"/~/media/mckinsey/email/shortlist/template/shortlist_promo-interrupter-849120982_1536x1536.jpg","alt":"the Shortlist"},"moreLinks":{"jsonValue":{"value":[{"CssClass":"","LinkType":"internal","Target":"","Text":"Get the Shortlist in your inbox","Title":"","Url":"/user-registration/manage-account/edit-subscriptions"}]}}},{"mediaID":{"value":"retailpodcast"},"title":{"value":"Subscribe to the \u003cem\u003eMcKinsey on Consumer and Retail\u003c/em\u003e podcast"},"description":{"value":""},"image":{"src":null,"alt":""},"moreLinks":{"jsonValue":{"value":[{"CssClass":"btn btn-fill","LinkType":"external","Target":"_blank","Text":"Apple Podcasts","Title":"","Url":"https://podcasts.apple.com/us/podcast/mckinsey-on-consumer-and-retail/id1526250428"},{"CssClass":"btn btn-fill","LinkType":"external","Target":"_blank","Text":"Spotify","Title":"","Url":"https://open.spotify.com/show/41D4bro8PPDOGhEgOEJyHB?si=7l3SeYCZQD-7v3toqd9fHQ"}]}}},{"mediaID":{"value":"insurancepodcast"},"title":{"value":"Want to subscribe to the \u003cem\u003eMcKinsey on Insurance\u003c/em\u003e podcast?"},"description":{"value":""},"image":{"src":null,"alt":""},"moreLinks":{"jsonValue":{"value":[{"CssClass":"btn btn-fill","LinkType":"external","Target":"_blank","Text":"Apple Podcasts","Title":"","Url":"https://podcasts.apple.com/us/podcast/mckinsey-on-insurance/id1533196359"},{"CssClass":"btn btn-fill","LinkType":"external","Target":"_blank","Text":"Spotify","Title":"","Url":"https://open.spotify.com/show/5gZ6YKF7n7C9zWaER4mFZ5"}]}}},{"mediaID":{"value":"strategypodcast"},"title":{"value":"Subscribe to the \u003cem\u003eInside the Strategy Room\u003c/em\u003e podcast"},"description":{"value":""},"image":{"src":null,"alt":""},"moreLinks":{"jsonValue":{"value":[{"CssClass":"btn btn-fill","LinkType":"external","Target":"_blank","Text":"Apple Podcasts","Title":"","Url":"https://itunes.apple.com/us/podcast/inside-the-strategy-room/id1422814215?mt=2"},{"CssClass":"btn btn-fill","LinkType":"external","Target":"","Text":"Spotify","Title":"","Url":"https://open.spotify.com/show/4TcFfiQ0e6OYuc5kRDLxqj"}]}}},{"mediaID":{"value":"mgipodcast"},"title":{"value":"Subscribe to the \u003cem\u003eForward Thinking\u003c/em\u003e podcast"},"description":{"value":""},"image":{"src":null,"alt":""},"moreLinks":{"jsonValue":{"value":[{"CssClass":"btn btn-fill","LinkType":"external","Target":"_blank","Text":"Apple Podcasts","Title":"","Url":"https://podcasts.apple.com/us/podcast/forward-thinking/id1319502839"},{"CssClass":"btn btn-fill","LinkType":"external","Target":"_blank","Text":"Spotify","Title":"","Url":"https://open.spotify.com/show/3S6fHLW6mcbDJDp1wkArKI?si=0zvPmiCsQ7SBAPanfpY4aA\u0026nd=1"},{"CssClass":"","LinkType":"internal","Target":"","Text":"Listen to previous episodes","Title":"","Url":"/mgi/forward-thinking"}]}}}]},"disruptorsignup":{"results":[{"mediaID":{"value":"full-width"},"renderMode":{"targetItem":{"key":{"value":"fullwidth"},"value":{"value":"Full Width"}}},"callToAction":{"url":"/user-registration/register-mvc","text":"Subscribe"},"title":{"jsonValue":{"value":"Stay current on your favorite topics"}}},{"mediaID":{"value":"left"},"renderMode":{"targetItem":{"key":{"value":"left"},"value":{"value":"Left"}}},"callToAction":{"url":"/user-registration/register-mvc","text":"Subscribe"},"title":{"jsonValue":{"value":"Stay current on your favorite topics"}}}]}},"ceros":{"results":[]},"exhibit":{"results":[]},"image":{"results":[{"mediaID":{"value":"get to know"},"description":{"value":"\u003ch5\u003eGet to know and directly engage with senior McKinsey experts on AI guardrails\u003c/h5\u003e\n\u003cp\u003e\u003cstrong\u003e\u003ca href=\"/our-people/lareina-yee\"\u003eLareina Yee\u003c/a\u003e\u003c/strong\u003e is a senior partner in McKinsey\u0026rsquo;s Bay Area office, where \u003cstrong\u003e\u003ca href=\"/our-people/roger-roberts\"\u003eRoger Roberts\u003c/a\u003e\u003c/strong\u003e is a partner; \u003cstrong\u003eMara Pometti\u003c/strong\u003e is a consultant in the London office; and \u003cstrong\u003eStephen Xu\u003c/strong\u003e is a senior director of product management in the Toronto office.\u003c/p\u003e"},"libraryMediaFile":{"src":null,"alt":""},"height":{"value":""},"width":{"value":""},"renderMode":{"targetItem":{"key":{"value":"Align Right"},"value":{"value":"Align Right"}}}},{"mediaID":{"value":"1"},"description":{"value":""},"libraryMediaFile":{"src":"/~/media/mckinsey/featured insights/mckinsey explainers/what are ai guardrails/what are ai guardrails_-2152543981-hero-1536x864.jpg","alt":"A pair of red and white concrete road barriers aligned on a street against a light blue background. "},"height":{"value":""},"width":{"value":""},"renderMode":{"targetItem":{"key":{"value":"Center"},"value":{"value":"Center"}}}}]},"exhibitcarousel":{"results":[]},"pullquote":{"results":[]},"disruptorsignup":{"results":[{"mediaID":{"value":"full-width"},"renderMode":{"targetItem":{"key":{"value":"fullwidth"},"value":{"value":"Full Width"}}},"callToAction":{"url":"/user-registration/register-mvc","text":"Subscribe"},"title":{"jsonValue":{"value":"Stay current on your favorite topics"}}},{"mediaID":{"value":"left"},"renderMode":{"targetItem":{"key":{"value":"left"},"value":{"value":"Left"}}},"callToAction":{"url":"/user-registration/register-mvc","text":"Subscribe"},"title":{"jsonValue":{"value":"Stay current on your favorite topics"}}}]},"articlesidebar":{"results":[{"mediaID":{"value":"Hype"},"description":{"jsonValue":{"value":"\u003cp\u003e\u003cstrong\u003eHyPe is a set\u003c/strong\u003e of technology-agnostic, adaptable components developed by McKinsey\u0026rsquo;s CustomerOne service line and powered by QuantumBlack, AI by McKinsey. Organizations can deploy HyPe components to accelerate hyperpersonalized customer engagements such as marketing messages. As part of the HyPe development process, we held workshops with multiple stakeholders to discuss the risks associated with hyperpersonalized content developed by AI as well as strategies to mitigate these risks. These workshops, along with analysis of the performance of the large language model used during the building and testing phases of HyPe, allowed us to create the guardrail categories described in this article. \u003c/p\u003e"}},"aboutTheAuthors":{"jsonValue":{"value":""}},"headline":{"jsonValue":{"value":"What is HyPe?"}},"footnotes":{"value":""},"showShareTools":{"boolValue":true},"backgroundColor":{"targetItem":null},"gradientDirection":{"targetItem":{"key":{"value":"Bottom Right"},"value":{"value":"bottom-right"}}},"renderMode":{"targetItem":{"key":{"value":"Collapsible"},"value":{"value":"default"}}},"isAboutAuthor":{"boolValue":false},"background":{"targetItem":null},"image":{"src":null,"alt":""}}]},"boxout":{"results":[]},"globalsidebar":{"results":[]},"video":{"results":[]},"scrollycontainer":{"results":[]},"bespokeinteractive":{"results":[]},"storycontainer":{"results":[]},"interactive":{"results":[]},"sectionheader":{"results":[]},"statementdisruptor":{"results":[]},"fullbleedphoto":{"results":[]},"tocitem":{"results":[]},"quizcontainer":{"results":[]},"generalup":{"results":[]},"twoupmedium":{"results":[]},"accordion":{"results":[]},"factoid":{"results":[]},"promobarwithquote":{"results":[]},"oneupmedium":{"results":[]},"oneupmediumquote":{"results":[]},"gridwall":{"results":[]},"twoupsmall":{"results":[]},"oneclicksubscribe":{"results":[]},"promobar":{"results":[]},"promobanner":{"results":[]},"sectionhero":{"results":[]},"threeuplinklist":{"results":[]},"table":{"results":[]},"explainertooltip":{"results":[]},"isFiveFiftyHorizontalArticle":{"boolValue":false},"fiveFifty":{"results":[]}},"contextItem":{"ancestors":[{"breadCrumbUrl":{"path":"/featured-insights/mckinsey-explainers"},"breadCrumbTitle":{"value":"McKinsey Explainers"},"isMiniSite":{"boolValue":false},"displayName":"McKinsey Explainers","template":{"id":"85FF05307883480F9A4C82123F72FFD8"}},{"breadCrumbUrl":{"path":"/featured-insights"},"breadCrumbTitle":{"value":"Explore our featured insights"},"isMiniSite":{"boolValue":false},"displayName":"Featured Insights","template":{"id":"85FF05307883480F9A4C82123F72FFD8"}},{"breadCrumbUrl":{"path":"/"},"breadCrumbTitle":{"value":""},"isMiniSite":{"boolValue":false},"displayName":"Home","template":{"id":"85FF05307883480F9A4C82123F72FFD8"}}]}}},"placeholders":{"main-area":[],"sidebar-area":[]}}]}},"itemId":"3899126a-e938-4641-bb09-fc15c3868982","pageEditing":false,"site":{"name":"website"},"pageState":"normal","language":"en","pageMetaData":{"alternateLanguages":[{"languageCode":"en","displayName":"English","url":"/featured-insights/mckinsey-explainers/what-are-ai-guardrails"}],"currentLanguage":"en","navigationLink":"featured-insights","activeItemId":"{3B4756D0-2A01-44CA-B37A-B4CA73D83717}","miniSiteId":"","officeCode":"","officeDisplayName":"","subscriptionPracticeData":null,"isAlaisedPage":false,"originalHostName":"www.mckinsey.com","updatedDate":"2024-11-18T18:15:19Z","createdDate":"2024-11-12T18:46:26Z","practice":{"isDefaultPractice":true,"name":"Monthly Highlights","code":"N02","stickyTitle":"Sign up for our Monthly Highlights newsletter","stickySubtitle":"Don't miss this roundup of our newest and most distinctive insights","previewLink":"\u003clink linktype=\"media\" target=\"\" id=\"{9D8859E5-868F-4618-9C77-073458526BA0}\" /\u003e","previewLinkLabel":null,"description":null,"descriptionText":null,"frequency":null,"regFormTitle":null},"url":"/featured-insights/mckinsey-explainers/what-are-ai-guardrails","isCareersPage":false,"officePath":"","hideAppPromo":false,"itemInfo":{"id":"3899126a-e938-4641-bb09-fc15c3868982","fullPath":"/sitecore/content/McKinsey/Home/Featured Insights/McKinsey Explainers/What are AI guardrails","templateName":"ArticleJSS","templateID":"683910db-02ba-40ba-92e7-726c880160a9"}},"defaultSettings":{"McKinsey Quarterly":"\u003cem\u003eMcKinsey Quarterly\u003c/em\u003e","FacebookCID":"soc-web","LinkedInCID":"soc-web","TwitterCID":"soc-web","Home Page SEO Details":{"Description":"McKinsey \u0026 Company is the trusted advisor and counselor to many of the world's most influential businesses and institutions.","Image_Url":"/~/media/images/global/seoimageplaceholder.jpg","Image_Alt":""}},"alumniDatalayerChunk":"static/chunks/alumnidatalayer.325ebb6069c1d69e.js"},"componentProps":{},"notFound":false,"noAccess":false},"__N_PREVIEW":true,"__N_SSP":true},"page":"/[[...path]]","query":{"timestamp":"1732564778869","path":["featured-insights","mckinsey-explainers","what-are-ai-guardrails"]},"buildId":"9vYT_fWQKRjfPT7lAZJdB","runtimeConfig":{"CONFIG_ENV":"production","CURRENT_ENV":"production","PUBLIC_URL":"https://www.mckinsey.com","PUBLIC_CMS_URL":"https://cms-prod.mckinsey.com","RECAPTCHA_V3_KEY":"6LdC5twSAAAAAF0dePIbY_ckeF05mKdYYJXn7uTg","INVISIBLE_RECAPTCHA_KEY":"6LcWCs0UAAAAAEik2NaGkfGH8mGHo1ThxIt-qUoE","USER_SERVICE_API_HOST":"https://prd-api.mckinsey.com","MOBILE_SERVICES_HOST":"https://mobileservices.mckinsey.com","SEARCH_SERVICE_API_HOST":"https://mckapi.mckinsey.com","OT_DATA_DOMAIN_SCRIPT_COM":"915b5091-0d7e-44d2-a8c4-cf08267e52fe","OT_DATA_DOMAIN_SCRIPT_DE":"13f50396-9d33-4d95-ba3e-1f108fa91009","OT_DATA_DOMAIN_SCRIPT_BR":"679eae24-1d5c-4f33-8aae-b35aa8269966","ADOBE_LAUNCH":"a400cfbb2fd3","BCOV_POLICY":"BCpkADawqM3Xn1B8iTmHawNytUSk0kXaQKMSM51CBVsYKHfwlRhZimTJqbOOq0NuyxTKQnvb4V9myMVRjwGkyGtJQO2HrUPOJHcsgkpbB-jyHwswonskf6XnZaF0TyyKkAQgbymSMtudeD9CLk-xCpqAzTaNUm7a3M3QKLN4pdLRNJ9_HbVMWzNpiUr5ezAVdMciDMwsD-gxNl43d7bfVP3wRhTNWbWUYn3WK7-Oz2gbxV6mMTm0jJU3SjK_x0meTprYUgdUWA-iDgwVglsnfJ_SVtVDPZypUokKv85GzFLM-ywjWlPM2Hg1ucemwwOpa-qvYBGfeMV7zP3n6s22hFCBveiWtHEBpV9N8f4rjmunMlxBx7kSlxIRjWquZm70hLRXMoxAgQ22BC86tBhgR_d1l9_-FzZuZIlZhzJwJi3ug1HszooHM8uEMOkHmB5nYj5XEgVk5JFYGcSwFSnMnNzE1ej2J3jcEiE3eSWHcremWjCZ0s9bs_8R08WpGWkBp8WMXUFfmhPBMk-MGixR_yIu7nJ6B_hYBnVhqKnnoW-xxEmRxgwG0Ev0S96cAcBG8qdlChGN7o1qN0heGTUIYnDk703q3lt9j7X4FgaITrlcPZOWZ96EHytVKUAtmtVp5BVGpiLn4viuUpEdAM_CLSBIaPVunk-39L8tm6m8iUYmbMX_KXPyaL900bQtVQgK39FCa0pUEZ_jSBOGNj1FH8NUPN2TS8_csNWfpN7gsaD9FBN3wUtawAD_RPdJ952odfwJqyvrkN89AyKnFdFMvUh6_4LCIQVYd4M5KC_k1WNLYb11KZhq6O4UaoyD4X1pV4-x0jBzEnc4VQTkL_04dw2xdPJ1uqUrmHuUcIBO0SKaEk6_5HYc3GKfYwmPz1oxMpOi-Rzw1QPyWIdbrcZdFbIHoVZ8FKfyslEv392pJMiKzUnVWL_3mZ0UqnGtddwlXYVdSs57AUdlrfOe7TVaaIMZ_kL7FSrbWrsaSufitSYr62xD1K1-qeb4wkx8QQEorxGJ3RG1Vpn97Pn5JEC_jvj0s7uzFgidYR9bp80EMmPCHUmRg-Hm1Aq1B4YMNc1ARnFz6UvT_8_pMhJEwKuypA-nPoJu4JZJNLVUdw","sitecoreApiKey":"{AD59BE63-BA20-47D4-A1B0-ECF30588E2E2}","sitecoreApiHost":"https://prod.mckinsey.com","FB_APP_ID":"1382278882079245","ARTICLE_API_HOST":"https://prd-api.mckinsey.com","SITECORE_FORMS_SERVICES_API_HOST":"https://prd-api.mckinsey.com","FEATURE_TOGGLES":"{\"stack_adapt_pixel\": true, \"article_seo_schema\":true, \"remember_me\": true, \"gql_promo_banner\": true, \"gql_highlight_bar\": true, \"gql_ceros_placeholder\": true, \"gql_two_up_medium\": true, \"gql_hero_with_partnership_mark\": true, \"gql_brightcove_gallery\": true, \"gql_one_up_medium\": true, \"gql_promo_bar_with_quote\": true, \"gql_fifty_fifty_hero\": true, \"gql_enhanched_hero\": true, \"gql_two_up_small\": true, \"gql_general_up\": true, \"gql_full_bleed_hero\": true, \"passwordless_login_enabled\": true, \"sso_enabled\": true, \"article_id\": \"\", \"personalisation_enabled\":true, \"hide_student_sticky_footer\":true, \"email_domain_validation_enabled\":true, \"enable_cid_cookie_qp\": true, \"otp_verification_enabled\": true}","HEADER_SCRIPT_HASH":"abcd1234","OKTA_AUTH_CLIENT_ID":"0oa8ppb3ypQ8uOI8y697","OKTA_AUTH_ISSUER":"https://dotcomidp.mckinsey.com/oauth2/aus2byav0jTElyFDD697/","AUTO_REDIRECTION_DURATION":"5","LINKEDIN_IDP":"0oaecyj1byxnfrfFk697","APPLE_IDP":"0oaeg2lxy6mq4PX5F697","GOOGLE_IDP":"0oaeczfkb03XL3E4O697","ALUMNI_BASE_API_URL":"https://dev-gateway.intranet.mckinsey.com/:apiKey/v1","ALUMNI_AUTHENTICATE_API_URL":"https://dev-ac3-okta-auth.intranet.mckinsey.com/authenticate","ALUMNI_OKTA_URL":"","ALUMNI_OKTA_ENABLED":"true","GOOGLE_MAP_API_KEY":"AIzaSyDqrGDNMDjy9pXHkMf-_Ut33LoC725GSe4","AWS_GATEWAY":"https://dev-gateway.mckinsey.com","AWS_API_GATEWAY_ID":"o63viscaj4","AWS_PERSONALIZATION_TRACKING_ID":"f3f97bc3-4ca7-4960-9e49-0bcff6294ab4","AWS_PERSONALIZATION_CID":"recommendations-onw-onw-mck-mck-art-2410-v1-user-ext-web","EVENTS_MODULE_CARDS_COUNT":"8","RECOMMENDATIONS_OFFER_ACTIVITY":"Article Recommendations"},"isFallback":false,"dynamicIds":[70281,15228,90390,76731,42354,35278,32991],"gssp":true,"locale":"en","locales":["en","da-DK"],"defaultLocale":"en","isPreview":true,"scriptLoader":[]}</script><script type="text/javascript" src="/4nI7/Dqwk/7B/nBsQ/ONiA/ariaDDVzf6XpO1aN/bTZxKC1lBA/FyIEUX/xKAk0"></script></body></html>