{"id":105916,"date":"2025-11-30T14:28:01","date_gmt":"2025-11-30T12:28:01","guid":{"rendered":"https:\/\/staging.checkmarx.com\/?p=105916"},"modified":"2025-12-28T08:10:33","modified_gmt":"2025-12-28T06:10:33","slug":"confronting-insecure-shadow-ai-six-must-have-capabilities","status":"publish","type":"post","link":"https:\/\/checkmarx.com\/blog\/confronting-insecure-shadow-ai-six-must-have-capabilities\/","title":{"rendered":"Confronting Insecure Shadow AI: Six Must-Have Capabilities"},"content":{"rendered":"<p>The speed of software delivery is no longer set by pipelines or processes; it\u2019s driven by prompts. Generative AI has transformed how code is created, shared, and deployed, dramatically improving developer productivity. Yet, visibility and governance haven\u2019t kept pace.&nbsp;<\/p>\n\n\n\n<p>Developers across enterprises are using GitHub Copilot, Cursor, and Replit AI to generate production code \u2013 often outside approved workflows. This invisible layer of AI-authored logic known as Shadow AI, untracked, AI-generated code entering production systems without policy enforcement or security validation.&nbsp;<\/p>\n\n\n\n<p>The problem isn\u2019t intent, it\u2019s infrastructure. Traditional AppSec tools were built for pipelines, not for prompts. They see only the <em>output<\/em> of the development process, never the <em>influence<\/em> of the assistant that helped shape it. To secure the AI-powered SDLC, organizations need a new kind of platform that\u2019s agentic, context-aware, and developer-native.&nbsp;<\/p>\n\n\n\n<h2 class=\"wp-block-heading article-anchor\" id=\"article-anchor-1\">The Shift From Reactive to Agentic AppSec&nbsp;<\/h2>\n\n\n\n<p>Legacy AppSec tools scan static artifacts long after code is written. Agentic AppSec tools, live inside the development experience.&nbsp;<\/p>\n\n\n\n<p>Agentic AppSec analyzes during the coding process and adapts to <em>developer intent<\/em>, enforcing <em>organizational policies<\/em> in real time. This process helps prevent insecure logic before it leaves the IDE and is pushed to production.&nbsp;<\/p>\n\n\n\n<p>The distinction is simple but profound:&nbsp;<\/p>\n\n\n\n<figure class=\"wp-block-table is-style-stripes\"><table class=\"has-fixed-layout\"><tbody>\n<tr>\n<td>\n<strong>Traditional AppSec<\/strong>&nbsp;<\/td>\n<td>\n<strong>Agentic AppSec<\/strong>&nbsp;<\/td>\n<\/tr>\n<tr>\n<td>Post-commit scanning (SAST, DAST, SCA)&nbsp;<\/td>\n<td>Pre-commit validation and guidance&nbsp;<\/td>\n<\/tr>\n<tr>\n<td>Operates on repositories&nbsp;<\/td>\n<td>Operates inside the IDE&nbsp;<\/td>\n<\/tr>\n<tr>\n<td>Detects known patterns&nbsp;<\/td>\n<td>Understands intent and origin&nbsp;<\/td>\n<\/tr>\n<tr>\n<td>Alerts after merge&nbsp;<\/td>\n<td>Prevents vulnerabilities before merge&nbsp;<\/td>\n<\/tr>\n<tr>\n<td>Static policies&nbsp;<\/td>\n<td>Context-adaptive governance&nbsp;<\/td>\n<\/tr>\n<\/tbody><\/table><\/figure>\n\n\n\n<p><\/p>\n\n\n\n<p><a href=\"https:\/\/checkmarx.com\/product\/developer-assist\/\" target=\"_blank\" rel=\"noreferrer noopener\">Checkmarx One Developer Assist<\/a>, powered by <a href=\"https:\/\/checkmarx.com\/glossary\/what-is-ai-code-security-assistance-acsa\/\" target=\"_blank\" rel=\"noreferrer noopener\">AI Code Security Assistance (ACSA)<\/a>, embodies this shift. Its developer-side agents analyze code as it\u2019s written (both human and AI-generated), providing inline fixes, safe refactors, and contextual reasoning without exposing source code outside the customer environment.&nbsp;<\/p>\n\n\n\n<h2 class=\"wp-block-heading article-anchor\" id=\"article-anchor-2\">Evaluating an Agentic AppSec Platform: Six Dimensions That Matter&nbsp;<\/h2>\n\n\n\n<p>Choosing the right platform means understanding what differentiates \u201cagentic\u201d from \u201cautomated.\u201d Below is a practical framework drawn from real Checkmarx deployments and independent buyer evaluations:&nbsp;<\/p>\n\n\n\n<h3 class=\"wp-block-heading\">1. Real-Time, Intent-Aware Validation&nbsp;<\/h3>\n\n\n\n<p>Agentic systems don\u2019t just parse syntax; they interpret intent.&nbsp;<\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li>Do they run continuously as developers write and modify code (not only at save or commit)? &nbsp;<\/li>\n\n\n\n<li>Can they correlate completions to assistant influence and block insecure logic inline? &nbsp;<\/li>\n\n\n\n<li>Do they explain why a fix is necessary, linking to policy, CVE, or data-flow context? &nbsp;<\/li>\n\n\n\n<li>Are unsafe suggestions from AI assistants intercepted before PR submission?&nbsp;<\/li>\n<\/ul>\n\n\n\n<p><strong>Example:<\/strong> Developer Assist recognizes when Copilot-generated code inserts an outdated encryption algorithm. It blocks the suggestion, explains the risk, and recommends a compliant alternative, all within the IDE. No context switching required.&nbsp;<\/p>\n\n\n\n<h3 class=\"wp-block-heading\">2. Developer-Centric UX and Trust&nbsp;<\/h3>\n\n\n\n<p>Adoption is critical. A technically strong tool that developers ignore provides zero ROI.&nbsp;<\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li>Is setup frictionless across IDEs like VS Code, JetBrains, Cursor, Windsurf, or Eclipse? &nbsp;<\/li>\n\n\n\n<li>Are results explainable, with clear diffs and one-click safe refactors? &nbsp;<\/li>\n\n\n\n<li>Can developers adjust noise levels, suppress false positives, or override with justification? &nbsp;<\/li>\n\n\n\n<li>Is latency low enough (&lt;200 ms feedback) to maintain flow?&nbsp;<\/li>\n<\/ul>\n\n\n\n<p>When developers realize that security can <em>accelerate<\/em> rather than interrupt their work, adoption skyrockets.&nbsp;<\/p>\n\n\n\n<h3 class=\"wp-block-heading\">3. Governance, Explainability, and Auditability&nbsp;<\/h3>\n\n\n\n<p>Agentic AppSec doesn&#8217;t governance, it embeds it instead.&nbsp;<\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li>Can roles, policies, and severities be defined per team, repo, or language? &nbsp;<\/li>\n\n\n\n<li>Are AI actions logged and explainable ( e.g. \u201cflagged due to unsafe deserialization pattern; see rule 143\u201d)? &nbsp;<\/li>\n\n\n\n<li>Can leaders audit overrides and monitor security drift over time? &nbsp;<\/li>\n\n\n\n<li>Does the system provide policy compliance dashboards for SOC 2, FedRAMP, or ISO 27001 mapping?&nbsp;<\/li>\n<\/ul>\n\n\n\n<p>Governance isn\u2019t a separate console anymore; it\u2019s a continuous feedback layer between the developer and the enterprise.&nbsp;<\/p>\n\n\n\n<h3 class=\"wp-block-heading\">4. Shadow AI Detection and Control&nbsp;<\/h3>\n\n\n\n<p>Every AI assistant represents a new integration surface and potential risk vector. <a href=\"https:\/\/checkmarx.com\/ai-llm-tools-in-application-security\/just-released-the-future-of-appsec-in-the-era-of-ai-2026-industry-outlook\/\" target=\"_blank\" rel=\"noreferrer noopener\">Shadow AI<\/a> occurs when developers use GenAI tools that generate or insert code outside sanctioned workflows. So even if the final code passes syntax checks, it may contain hidden dependencies, unvetted packages, or logic trained on insecure repositories.&nbsp;<\/p>\n\n\n\n<p>Key capabilities to demand:&nbsp;<\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li>\n<strong>Detection:<\/strong> Identify AI-authored snippets by token pattern, prompt signature, or model fingerprint.&nbsp;<\/li>\n\n\n\n<li>\n<strong>Attribution:<\/strong> Map completions to the tool of origin (Copilot, Replit, Cursor, Windsurf).&nbsp;<\/li>\n\n\n\n<li>\n<strong>Risk Scoring:<\/strong> Flag AI-influenced logic that bypasses review or policy validation.&nbsp;<\/li>\n\n\n\n<li>\n<strong>Policy Enforcement:<\/strong> Block commits from unapproved assistants or require inline re-validation.&nbsp;<\/li>\n\n\n\n<li>\n<strong>Reporting:<\/strong> Provide dashboards showing AI usage by team, repo, or project.&nbsp;<\/li>\n<\/ul>\n\n\n\n<p>Given the ubiquitous adoption of Gen AI coding practices, shadow AI isn&#8217;t a hypothetical risk anymore.&nbsp;<\/p>\n\n\n\n<h3 class=\"wp-block-heading\">5. ROI and Throughput Gains&nbsp;<\/h3>\n\n\n\n<p>Agentic AppSec doesn\u2019t just shift <em>when<\/em> vulnerabilities are found; it also changes <em>how much<\/em> they cost to fix.&nbsp;<\/p>\n\n\n\n<p>According to Checkmarx\u2019s internal ROI analysis (2025):&nbsp;<\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li>\n<strong>MTTR improved by 30\u201340%<\/strong> with inline remediation versus post-merge fixes.&nbsp;<\/li>\n\n\n\n<li>\n<strong>Development throughput increased by 20\u201325%<\/strong> due to fewer broken builds and CI\/CD reruns.&nbsp;<\/li>\n\n\n\n<li>\n<strong>Cost-per-vulnerability dropped by 35%<\/strong>, with early detection eliminating redundant rework cycles.&nbsp;<\/li>\n\n\n\n<li>\n<strong>Safe Refactor<\/strong> capabilities cut dependency-upgrade effort by up to <strong>60\u201370%<\/strong>, reducing technical debt at scale.&nbsp;<\/li>\n<\/ul>\n\n\n\n<p>These metrics correlate directly with <a href=\"https:\/\/checkmarx.com\/blog\/tuning-appsec-to-boost-your-dora-metrics\/\" target=\"_blank\" rel=\"noreferrer noopener\">improved DORA outcomes<\/a>, including faster lead time for changes, reduced change-failure rate, and higher deployment frequency.&nbsp;<\/p>\n\n\n\n<h3 class=\"wp-block-heading\">6. Ecosystem and Integration Fit&nbsp;<\/h3>\n\n\n\n<p>No AI agent operates in isolation. An effective platform must connect seamlessly across your engineering stack:&nbsp;<\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li>\n<strong>IDEs:<\/strong> VS Code, JetBrains, Cursor, Eclipse, Windsurf.&nbsp;<\/li>\n\n\n\n<li>\n<strong>Version Control:<\/strong> GitHub, GitLab, Bitbucket.&nbsp;<\/li>\n\n\n\n<li>\n<strong>CI\/CD:<\/strong> Jenkins, Azure DevOps, CircleCI, GitHub Actions.&nbsp;<\/li>\n\n\n\n<li>\n<strong>Package Managers:<\/strong> npm, PyPI, Maven, and Go modules with real-time SCA policy checks.&nbsp;<\/li>\n\n\n\n<li>\n<strong>SIEM\/SOAR:<\/strong> Splunk, ServiceNow for alert ingestion and incident correlation.&nbsp;<\/li>\n<\/ul>\n\n\n\n<p>Checkmarx\u2019s open APIs enable these integrations while maintaining strict data sovereignty. No source code leaves the customer\u2019s environment.\u00a0<\/p>\n\n\n\n<p>These six dimensions align closely with the broader evaluation framework outlined in <a href=\"https:\/\/checkmarx.com\/the-agentic-ai-buyers-guide\/\"><strong>The Agentic AI Buyer\u2019s Guide<\/strong><\/a>, which covers Agentic AppSec, Shadow AI governance, and AI Code Security Assistance (ACSA) in depth. This free guide helps benchmark vendors, capabilities, and ROI expectations when adopting AI.<\/p>\n\n\n\n<h2 class=\"wp-block-heading article-anchor\" id=\"article-anchor-3\">The Shadow AI Reality: Unseen Code, Unscanned Risk&nbsp;<\/h2>\n\n\n\n<p>Picture this scenario: a backend developer experimenting with Cursor generates a new authentication handler. Cursor auto-imports an outdated JSON-web-token package containing a known CVE. Because the commit passes linting and functional tests, it merges successfully, but the vulnerability isn\u2019t caught until weeks later, when CI\/CD scanning reveals it post-deployment.&nbsp;&nbsp;<\/p>\n\n\n\n<p>That\u2019s the shadow AI gap. Developers weren\u2019t careless &#8211; the tooling chain wasn\u2019t built to recognize intent or origin. Agentic AppSec platforms close that gap by embedding reasoning at the moment of creation &#8211; before commit, before merge, before any exposure.&nbsp;<\/p>\n\n\n\n<h2 class=\"wp-block-heading article-anchor\" id=\"article-anchor-4\">Building an Evaluation Shortlist&nbsp;<\/h2>\n\n\n\n<p>When comparing vendors, prioritize the following questions:&nbsp;<\/p>\n\n\n\n<ol class=\"wp-block-list\">\n<li>Does the platform operate natively within the IDE and correlate assistant influence? &nbsp;<\/li>\n\n\n\n<li>Can it enforce pre-commit policy gates without sending code externally? &nbsp;<\/li>\n\n\n\n<li>Does it quantify throughput and MTTR gains with customer-verified data? &nbsp;<\/li>\n\n\n\n<li>Is explainability built in, can every decision be traced and justified?&nbsp;<\/li>\n<\/ol>\n\n\n\n<p>Ask for proof, not promises \u2013 real customer metrics, not theoretical benchmarks.&nbsp;<\/p>\n\n\n\n<h2 class=\"wp-block-heading article-anchor\" id=\"article-anchor-5\">The Business Case: Why It Matters Now&nbsp;<\/h2>\n\n\n\n<p>The economics of software delivery are shifting fast. AI has removed the bottleneck of creation, but not the cost of correction. Every vulnerability found after commit costs exponentially more to fix, and the gap only widens with each assistant-authored line of code.&nbsp;<\/p>\n\n\n\n<p>By shifting validation left of commit, agentic AppSec platforms deliver measurable ROI:&nbsp;<\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li>\n<strong>Security Debt Reduction<\/strong>: Early prevention reduces accumulated risk. &nbsp;<\/li>\n\n\n\n<li>\n<strong>Velocity Retention<\/strong>: Inline fixes avoid blocking developers mid-flow. &nbsp;<\/li>\n\n\n\n<li>\n<strong>Regulatory Alignment<\/strong>: AI governance satisfies evolving compliance mandates. &nbsp;<\/li>\n\n\n\n<li>\n<strong>Cross-Team Synergy<\/strong>: Security, DevOps, and compliance work from shared telemetry.&nbsp;<\/li>\n<\/ul>\n\n\n\n<p>In other words, security finally scales with speed.&nbsp;<\/p>\n\n\n\n<h2 class=\"wp-block-heading article-anchor\" id=\"article-anchor-6\">Closing the Loop: Visibility, Velocity, and Verification&nbsp;<\/h2>\n\n\n\n<p>Shadow AI isn\u2019t going away. If anything, the next generation of AI assistants will be more autonomous, creative, and capable of introducing even more subtle vulnerabilities that bypass traditional defenses.&nbsp;<\/p>\n\n\n\n<p>Agentic AppSec turns risk into resilience. By validating intent, governing policy, and embedding reasoning directly inside the developer\u2019s workspace, platforms like <a href=\"https:\/\/checkmarx.com\/product\/checkmarx-one-assist\/\" target=\"_blank\" rel=\"noreferrer noopener\">Checkmarx One Assist<\/a> transform AppSec from a reactive gate into a proactive guide.&nbsp;<\/p>\n\n\n\n<p><strong>The result:<\/strong> fewer vulnerabilities, faster releases, and a measurable reduction in AppSec overhead without slowing innovation.&nbsp;<\/p>\n\n\n\n<p><strong>Next in the series:<\/strong> <a href=\"https:\/\/checkmarx.com\/blog\/the-agentic-future-of-appsec-measuring-impact-and-securing-the-ai-powered-sdlc\/\">Measuring Impact and Securing the AI-Powered SDLC<\/a>.<\/p>","protected":false},"excerpt":{"rendered":"<p>The speed of software delivery is no longer set by pipelines or processes; it\u2019s driven by prompts. Generative AI has transformed how code is created, shared, and deployed, dramatically improving developer productivity. Yet, visibility and governance haven\u2019t kept pace.&nbsp; Developers across enterprises are using GitHub Copilot, Cursor, and Replit AI to generate production code \u2013 [&hellip;]<\/p>\n","protected":false},"author":32,"featured_media":105917,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"_acf_changed":false,"footnotes":""},"categories":[1284,85,84],"tags":[1272,1452,1410],"class_list":["post-105916","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-ai-llm-tools-in-application-security","category-application-security-trends","category-blog","tag-agentic-ai","tag-developer-assist","tag-genai-security"],"acf":[],"yoast_head":"<!-- This site is optimized with the Yoast SEO plugin v27.1.1 - https:\/\/yoast.com\/product\/yoast-seo-wordpress\/ -->\n<title>Confronting Insecure Shadow AI: Six Must-Have Capabilities<\/title>\n<meta name=\"description\" content=\"To secure the AI-powered SDLC, organizations need a new kind of platform that\u2019s agentic, context-aware, and developer-native.\" \/>\n<meta name=\"robots\" content=\"index, follow, max-snippet:-1, max-image-preview:large, max-video-preview:-1\" \/>\n<link rel=\"canonical\" href=\"https:\/\/checkmarx.com\/blog\/confronting-insecure-shadow-ai-six-must-have-capabilities\/\" \/>\n<meta property=\"og:locale\" content=\"en_US\" \/>\n<meta property=\"og:type\" content=\"article\" \/>\n<meta property=\"og:title\" content=\"Confronting Insecure Shadow AI: Six Must-Have Capabilities\" \/>\n<meta property=\"og:description\" content=\"To secure the AI-powered SDLC, organizations need a new kind of platform that\u2019s agentic, context-aware, and developer-native.\" \/>\n<meta property=\"og:url\" content=\"https:\/\/checkmarx.com\/blog\/confronting-insecure-shadow-ai-six-must-have-capabilities\/\" \/>\n<meta property=\"og:site_name\" content=\"Checkmarx\" \/>\n<meta property=\"article:publisher\" content=\"https:\/\/www.facebook.com\/Checkmarx.Source.Code.Analysis\" \/>\n<meta property=\"article:published_time\" content=\"2025-11-30T12:28:01+00:00\" \/>\n<meta property=\"article:modified_time\" content=\"2025-12-28T06:10:33+00:00\" \/>\n<meta property=\"og:image\" content=\"https:\/\/checkmarx.com\/wp-content\/uploads\/2025\/11\/Blog-Banner-pt-3.png\" \/>\n\t<meta property=\"og:image:width\" content=\"2560\" \/>\n\t<meta property=\"og:image:height\" content=\"1280\" \/>\n\t<meta property=\"og:image:type\" content=\"image\/png\" \/>\n<meta name=\"author\" content=\"Rebecca Spiegel\" \/>\n<meta name=\"twitter:card\" content=\"summary_large_image\" \/>\n<meta name=\"twitter:creator\" content=\"@checkmarx\" \/>\n<meta name=\"twitter:site\" content=\"@checkmarx\" \/>\n<meta name=\"twitter:label1\" content=\"Written by\" \/>\n\t<meta name=\"twitter:data1\" content=\"Rebecca Spiegel\" \/>\n\t<meta name=\"twitter:label2\" content=\"Est. reading time\" \/>\n\t<meta name=\"twitter:data2\" content=\"7 minutes\" \/>\n<script type=\"application\/ld+json\" class=\"yoast-schema-graph\">{\"@context\":\"https:\/\/schema.org\",\"@graph\":[{\"@type\":\"Article\",\"@id\":\"https:\/\/checkmarx.com\/blog\/confronting-insecure-shadow-ai-six-must-have-capabilities\/#article\",\"isPartOf\":{\"@id\":\"https:\/\/checkmarx.com\/blog\/confronting-insecure-shadow-ai-six-must-have-capabilities\/\"},\"author\":{\"name\":\"Rebecca Spiegel\",\"@id\":\"https:\/\/checkmarx.com\/#\/schema\/person\/a3ab30b42e891e3562aa46a70bbb0674\"},\"headline\":\"Confronting Insecure Shadow AI: Six Must-Have Capabilities\",\"datePublished\":\"2025-11-30T12:28:01+00:00\",\"dateModified\":\"2025-12-28T06:10:33+00:00\",\"mainEntityOfPage\":{\"@id\":\"https:\/\/checkmarx.com\/blog\/confronting-insecure-shadow-ai-six-must-have-capabilities\/\"},\"wordCount\":1402,\"publisher\":{\"@id\":\"https:\/\/checkmarx.com\/#organization\"},\"image\":{\"@id\":\"https:\/\/checkmarx.com\/blog\/confronting-insecure-shadow-ai-six-must-have-capabilities\/#primaryimage\"},\"thumbnailUrl\":\"https:\/\/checkmarx.com\/wp-content\/uploads\/2025\/11\/Blog-Banner-pt-3.png\",\"keywords\":[\"Agentic AI\",\"developer assist\",\"GenAI Security\"],\"articleSection\":[\"AI &amp; LLM Tools in Application Security\",\"Application Security Trends &amp; Insights\",\"Blog\"],\"inLanguage\":\"en-US\"},{\"@type\":\"WebPage\",\"@id\":\"https:\/\/checkmarx.com\/blog\/confronting-insecure-shadow-ai-six-must-have-capabilities\/\",\"url\":\"https:\/\/checkmarx.com\/blog\/confronting-insecure-shadow-ai-six-must-have-capabilities\/\",\"name\":\"Confronting Insecure Shadow AI: Six Must-Have Capabilities\",\"isPartOf\":{\"@id\":\"https:\/\/checkmarx.com\/#website\"},\"primaryImageOfPage\":{\"@id\":\"https:\/\/checkmarx.com\/blog\/confronting-insecure-shadow-ai-six-must-have-capabilities\/#primaryimage\"},\"image\":{\"@id\":\"https:\/\/checkmarx.com\/blog\/confronting-insecure-shadow-ai-six-must-have-capabilities\/#primaryimage\"},\"thumbnailUrl\":\"https:\/\/checkmarx.com\/wp-content\/uploads\/2025\/11\/Blog-Banner-pt-3.png\",\"datePublished\":\"2025-11-30T12:28:01+00:00\",\"dateModified\":\"2025-12-28T06:10:33+00:00\",\"description\":\"To secure the AI-powered SDLC, organizations need a new kind of platform that\u2019s agentic, context-aware, and developer-native.\",\"inLanguage\":\"en-US\",\"potentialAction\":[{\"@type\":\"ReadAction\",\"target\":[\"https:\/\/checkmarx.com\/blog\/confronting-insecure-shadow-ai-six-must-have-capabilities\/\"]}]},{\"@type\":\"ImageObject\",\"inLanguage\":\"en-US\",\"@id\":\"https:\/\/checkmarx.com\/blog\/confronting-insecure-shadow-ai-six-must-have-capabilities\/#primaryimage\",\"url\":\"https:\/\/checkmarx.com\/wp-content\/uploads\/2025\/11\/Blog-Banner-pt-3.png\",\"contentUrl\":\"https:\/\/checkmarx.com\/wp-content\/uploads\/2025\/11\/Blog-Banner-pt-3.png\",\"width\":2560,\"height\":1280,\"caption\":\"Confronting Insecure Shadow AI\"},{\"@type\":\"WebSite\",\"@id\":\"https:\/\/checkmarx.com\/#website\",\"url\":\"https:\/\/checkmarx.com\/\",\"name\":\"Checkmarx\",\"description\":\"The world runs on code. We secure it.\",\"publisher\":{\"@id\":\"https:\/\/checkmarx.com\/#organization\"},\"potentialAction\":[{\"@type\":\"SearchAction\",\"target\":{\"@type\":\"EntryPoint\",\"urlTemplate\":\"https:\/\/checkmarx.com\/?s={search_term_string}\"},\"query-input\":{\"@type\":\"PropertyValueSpecification\",\"valueRequired\":true,\"valueName\":\"search_term_string\"}}],\"inLanguage\":\"en-US\"},{\"@type\":\"Organization\",\"@id\":\"https:\/\/checkmarx.com\/#organization\",\"name\":\"Checkmarx\",\"url\":\"https:\/\/checkmarx.com\/\",\"logo\":{\"@type\":\"ImageObject\",\"inLanguage\":\"en-US\",\"@id\":\"https:\/\/checkmarx.com\/#\/schema\/logo\/image\/\",\"url\":\"https:\/\/checkmarx.com\/wp-content\/uploads\/2024\/02\/logo-dark.svg\",\"contentUrl\":\"https:\/\/checkmarx.com\/wp-content\/uploads\/2024\/02\/logo-dark.svg\",\"width\":1,\"height\":1,\"caption\":\"Checkmarx\"},\"image\":{\"@id\":\"https:\/\/checkmarx.com\/#\/schema\/logo\/image\/\"},\"sameAs\":[\"https:\/\/www.facebook.com\/Checkmarx.Source.Code.Analysis\",\"https:\/\/x.com\/checkmarx\",\"https:\/\/www.youtube.com\/user\/CheckmarxResearchLab\",\"https:\/\/www.linkedin.com\/company\/checkmarx\"]},{\"@type\":\"Person\",\"@id\":\"https:\/\/checkmarx.com\/#\/schema\/person\/a3ab30b42e891e3562aa46a70bbb0674\",\"name\":\"Rebecca Spiegel\",\"image\":{\"@type\":\"ImageObject\",\"inLanguage\":\"en-US\",\"@id\":\"https:\/\/checkmarx.com\/#\/schema\/person\/image\/\",\"url\":\"https:\/\/checkmarx.com\/wp-content\/uploads\/2024\/06\/avatar_32.jpg\",\"contentUrl\":\"https:\/\/checkmarx.com\/wp-content\/uploads\/2024\/06\/avatar_32.jpg\",\"caption\":\"Rebecca Spiegel\"},\"url\":\"https:\/\/checkmarx.com\/author\/rebecca\/\"}]}<\/script>\n<!-- \/ Yoast SEO plugin. -->","yoast_head_json":{"title":"Confronting Insecure Shadow AI: Six Must-Have Capabilities","description":"To secure the AI-powered SDLC, organizations need a new kind of platform that\u2019s agentic, context-aware, and developer-native.","robots":{"index":"index","follow":"follow","max-snippet":"max-snippet:-1","max-image-preview":"max-image-preview:large","max-video-preview":"max-video-preview:-1"},"canonical":"https:\/\/checkmarx.com\/blog\/confronting-insecure-shadow-ai-six-must-have-capabilities\/","og_locale":"en_US","og_type":"article","og_title":"Confronting Insecure Shadow AI: Six Must-Have Capabilities","og_description":"To secure the AI-powered SDLC, organizations need a new kind of platform that\u2019s agentic, context-aware, and developer-native.","og_url":"https:\/\/checkmarx.com\/blog\/confronting-insecure-shadow-ai-six-must-have-capabilities\/","og_site_name":"Checkmarx","article_publisher":"https:\/\/www.facebook.com\/Checkmarx.Source.Code.Analysis","article_published_time":"2025-11-30T12:28:01+00:00","article_modified_time":"2025-12-28T06:10:33+00:00","og_image":[{"width":2560,"height":1280,"url":"https:\/\/checkmarx.com\/wp-content\/uploads\/2025\/11\/Blog-Banner-pt-3.png","type":"image\/png"}],"author":"Rebecca Spiegel","twitter_card":"summary_large_image","twitter_creator":"@checkmarx","twitter_site":"@checkmarx","twitter_misc":{"Written by":"Rebecca Spiegel","Est. reading time":"7 minutes"},"schema":{"@context":"https:\/\/schema.org","@graph":[{"@type":"Article","@id":"https:\/\/checkmarx.com\/blog\/confronting-insecure-shadow-ai-six-must-have-capabilities\/#article","isPartOf":{"@id":"https:\/\/checkmarx.com\/blog\/confronting-insecure-shadow-ai-six-must-have-capabilities\/"},"author":{"name":"Rebecca Spiegel","@id":"https:\/\/checkmarx.com\/#\/schema\/person\/a3ab30b42e891e3562aa46a70bbb0674"},"headline":"Confronting Insecure Shadow AI: Six Must-Have Capabilities","datePublished":"2025-11-30T12:28:01+00:00","dateModified":"2025-12-28T06:10:33+00:00","mainEntityOfPage":{"@id":"https:\/\/checkmarx.com\/blog\/confronting-insecure-shadow-ai-six-must-have-capabilities\/"},"wordCount":1402,"publisher":{"@id":"https:\/\/checkmarx.com\/#organization"},"image":{"@id":"https:\/\/checkmarx.com\/blog\/confronting-insecure-shadow-ai-six-must-have-capabilities\/#primaryimage"},"thumbnailUrl":"https:\/\/checkmarx.com\/wp-content\/uploads\/2025\/11\/Blog-Banner-pt-3.png","keywords":["Agentic AI","developer assist","GenAI Security"],"articleSection":["AI &amp; LLM Tools in Application Security","Application Security Trends &amp; Insights","Blog"],"inLanguage":"en-US"},{"@type":"WebPage","@id":"https:\/\/checkmarx.com\/blog\/confronting-insecure-shadow-ai-six-must-have-capabilities\/","url":"https:\/\/checkmarx.com\/blog\/confronting-insecure-shadow-ai-six-must-have-capabilities\/","name":"Confronting Insecure Shadow AI: Six Must-Have Capabilities","isPartOf":{"@id":"https:\/\/checkmarx.com\/#website"},"primaryImageOfPage":{"@id":"https:\/\/checkmarx.com\/blog\/confronting-insecure-shadow-ai-six-must-have-capabilities\/#primaryimage"},"image":{"@id":"https:\/\/checkmarx.com\/blog\/confronting-insecure-shadow-ai-six-must-have-capabilities\/#primaryimage"},"thumbnailUrl":"https:\/\/checkmarx.com\/wp-content\/uploads\/2025\/11\/Blog-Banner-pt-3.png","datePublished":"2025-11-30T12:28:01+00:00","dateModified":"2025-12-28T06:10:33+00:00","description":"To secure the AI-powered SDLC, organizations need a new kind of platform that\u2019s agentic, context-aware, and developer-native.","inLanguage":"en-US","potentialAction":[{"@type":"ReadAction","target":["https:\/\/checkmarx.com\/blog\/confronting-insecure-shadow-ai-six-must-have-capabilities\/"]}]},{"@type":"ImageObject","inLanguage":"en-US","@id":"https:\/\/checkmarx.com\/blog\/confronting-insecure-shadow-ai-six-must-have-capabilities\/#primaryimage","url":"https:\/\/checkmarx.com\/wp-content\/uploads\/2025\/11\/Blog-Banner-pt-3.png","contentUrl":"https:\/\/checkmarx.com\/wp-content\/uploads\/2025\/11\/Blog-Banner-pt-3.png","width":2560,"height":1280,"caption":"Confronting Insecure Shadow AI"},{"@type":"WebSite","@id":"https:\/\/checkmarx.com\/#website","url":"https:\/\/checkmarx.com\/","name":"Checkmarx","description":"The world runs on code. We secure it.","publisher":{"@id":"https:\/\/checkmarx.com\/#organization"},"potentialAction":[{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https:\/\/checkmarx.com\/?s={search_term_string}"},"query-input":{"@type":"PropertyValueSpecification","valueRequired":true,"valueName":"search_term_string"}}],"inLanguage":"en-US"},{"@type":"Organization","@id":"https:\/\/checkmarx.com\/#organization","name":"Checkmarx","url":"https:\/\/checkmarx.com\/","logo":{"@type":"ImageObject","inLanguage":"en-US","@id":"https:\/\/checkmarx.com\/#\/schema\/logo\/image\/","url":"https:\/\/checkmarx.com\/wp-content\/uploads\/2024\/02\/logo-dark.svg","contentUrl":"https:\/\/checkmarx.com\/wp-content\/uploads\/2024\/02\/logo-dark.svg","width":1,"height":1,"caption":"Checkmarx"},"image":{"@id":"https:\/\/checkmarx.com\/#\/schema\/logo\/image\/"},"sameAs":["https:\/\/www.facebook.com\/Checkmarx.Source.Code.Analysis","https:\/\/x.com\/checkmarx","https:\/\/www.youtube.com\/user\/CheckmarxResearchLab","https:\/\/www.linkedin.com\/company\/checkmarx"]},{"@type":"Person","@id":"https:\/\/checkmarx.com\/#\/schema\/person\/a3ab30b42e891e3562aa46a70bbb0674","name":"Rebecca Spiegel","image":{"@type":"ImageObject","inLanguage":"en-US","@id":"https:\/\/checkmarx.com\/#\/schema\/person\/image\/","url":"https:\/\/checkmarx.com\/wp-content\/uploads\/2024\/06\/avatar_32.jpg","contentUrl":"https:\/\/checkmarx.com\/wp-content\/uploads\/2024\/06\/avatar_32.jpg","caption":"Rebecca Spiegel"},"url":"https:\/\/checkmarx.com\/author\/rebecca\/"}]}},"_links":{"self":[{"href":"https:\/\/checkmarx.com\/wp-json\/wp\/v2\/posts\/105916","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/checkmarx.com\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/checkmarx.com\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/checkmarx.com\/wp-json\/wp\/v2\/users\/32"}],"replies":[{"embeddable":true,"href":"https:\/\/checkmarx.com\/wp-json\/wp\/v2\/comments?post=105916"}],"version-history":[{"count":0,"href":"https:\/\/checkmarx.com\/wp-json\/wp\/v2\/posts\/105916\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/checkmarx.com\/wp-json\/wp\/v2\/media\/105917"}],"wp:attachment":[{"href":"https:\/\/checkmarx.com\/wp-json\/wp\/v2\/media?parent=105916"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/checkmarx.com\/wp-json\/wp\/v2\/categories?post=105916"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/checkmarx.com\/wp-json\/wp\/v2\/tags?post=105916"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}