{"id":6889,"date":"2025-12-22T22:42:01","date_gmt":"2025-12-22T22:42:01","guid":{"rendered":"https:\/\/h-in-q.com\/?p=6889"},"modified":"2025-12-22T22:42:01","modified_gmt":"2025-12-22T22:42:01","slug":"ai-is-not-a-chatbot-why-95-percent-of-projects-fail","status":"publish","type":"post","link":"https:\/\/h-in-q.com\/fr\/blog\/ai-is-not-a-chatbot-why-95-percent-of-projects-fail\/","title":{"rendered":"AI Is Not a Chatbot: The Real Reason 95% of AI Projects Fail"},"content":{"rendered":"<h2>The AI Illusion That\u2019s Crippling Modern Business<\/h2>\n<p>We are living through one of the most dangerous illusions in corporate technology today: the belief that a chatbot subscription equals an AI solution. Too many executives think that \u201cAI imagine\u201d means buying ChatGPT and layering a conversational interface on a simple use case. According to research tied to an <strong>MIT study revealing that ~95% of enterprise generative AI initiatives fail to deliver measurable business impact<\/strong>, this misinterpretation is costly and widespread. (<a href=\"https:\/\/www.tomshardware.com\/tech-industry\/artificial-intelligence\/95-percent-of-generative-ai-implementations-in-enterprise-have-no-measurable-impact-on-p-and-l-says-mit-flawed-integration-key-reason-why-ai-projects-underperform?utm_source=chatgpt.com\" target=\"_blank\" rel=\"noopener\">Tom&#8217;s Hardware<\/a>) Real AI is not a widget you drop into Microsoft Teams or Salesforce. It must be architected as a living platform that intertwines <strong>data readiness<\/strong>, workflows, security, integration, and culture; exactly the kinds of systemic capabilities that distinguish the few successful implementations from the many failures. (<a href=\"https:\/\/www.mckinsey.com\/capabilities\/tech-and-ai\/our-insights\/charting-a-path-to-the-data-and-ai-driven-enterprise-of-2030?utm_source=chatgpt.com\" target=\"_blank\" rel=\"noopener\">McKinsey &amp; Company<\/a>) This article dismantles the illusion and shows why superficial AI leads to failure, while systemic AI drives value.<\/p>\n<h2>The Grand Illusion: Why AI \u2260 Chatbot<\/h2>\n<h3>Hype vs. Reality: The Chatbot Misconception<\/h3>\n<p>There\u2019s a gulf between expectation and reality. Many companies treat AI like a <strong>chatbot subscription<\/strong>, licensing tools without grounding them in business problems or platforms. Generic AI tools may seem enticing, but superficial deployments fail to integrate with core systems, leading to the very stagnant results cited in enterprise failure research. (<a href=\"https:\/\/mlq.ai\/media\/quarterly_decks\/v0.1_State_of_AI_in_Business_2025_Report.pdf?utm_source=chatgpt.com\" target=\"_blank\" rel=\"noopener\">mlq.ai<\/a>)<\/p>\n<h3>How Corporate Hype Warps AI Strategy<\/h3>\n<p>This hype isn\u2019t organic; it\u2019s reinforced by vendors and consultants who profit from selling quick solutions. Boards see the marketing narratives and assume implementation will be equally simple, only to find that <strong>enterprise AI requires complex engineering and organizational alignment<\/strong>. (<a href=\"https:\/\/www.mckinsey.com\/capabilities\/tech-and-ai\/our-insights\/charting-a-path-to-the-data-and-ai-driven-enterprise-of-2030?utm_source=chatgpt.com\" target=\"_blank\" rel=\"noopener\">McKinsey &amp; Company<\/a>)<\/p>\n<h3>Common Enterprise AI Misconceptions<\/h3>\n<table style=\"height: 16px;\" width=\"5\">\n<thead>\n<tr>\n<td><img loading=\"lazy\" decoding=\"async\" class=\"aligncenter size-medium wp-image-6890\" src=\"https:\/\/h-in-q.com\/wp-content\/uploads\/2025\/12\/misconception-of-reality-640x426.jpg\" alt=\"Common Enterprise AI Misconceptions\" width=\"640\" height=\"426\" srcset=\"https:\/\/h-in-q.com\/wp-content\/uploads\/2025\/12\/misconception-of-reality-640x426.jpg 640w, https:\/\/h-in-q.com\/wp-content\/uploads\/2025\/12\/misconception-of-reality-768x512.jpg 768w, https:\/\/h-in-q.com\/wp-content\/uploads\/2025\/12\/misconception-of-reality-320x213.jpg 320w, https:\/\/h-in-q.com\/wp-content\/uploads\/2025\/12\/misconception-of-reality.jpg 1211w\" sizes=\"auto, (max-width: 640px) 100vw, 640px\" \/><\/td>\n<td><\/td>\n<\/tr>\n<\/thead>\n<\/table>\n<h2>2. The Hidden Reality: 95% of AI Isn\u2019t \u201cAI\u201d<\/h2>\n<h3>Software Engineering: The Real Backbone<\/h3>\n<p>Despite popular narratives, <strong>95% of the work in AI solutions is not about machine reasoning<\/strong>. It\u2019s about software engineering, data orchestration, and lifecycle management; what industry experts classify as infrastructure and governance. (<a href=\"https:\/\/www.mckinsey.com\/capabilities\/tech-and-ai\/our-insights\/charting-a-path-to-the-data-and-ai-driven-enterprise-of-2030?utm_source=chatgpt.com\" target=\"_blank\" rel=\"noopener\">McKinsey &amp; Company<\/a>) AI models excel at reasoning, but a platform <em>turns reasoning into business impact<\/em>.<\/p>\n<h3>From Orchestration to Monitoring<\/h3>\n<p>AI doesn\u2019t exist in a bubble. To be trustworthy and resilient, AI systems need orchestration layers that handle dependencies, scaling, and failure. Reliable monitoring dashboards visualize latency, data drift, and throughput in real time essential for business continuity.<\/p>\n<h3>Core Engineering Components<\/h3>\n<p><img loading=\"lazy\" decoding=\"async\" class=\"aligncenter size-medium wp-image-6891\" src=\"https:\/\/h-in-q.com\/wp-content\/uploads\/2025\/12\/Backend-infrastructure-diagram-with-pipelines-640x427.webp\" alt=\"Backend infrastructure diagram with pipelines\" width=\"640\" height=\"427\" srcset=\"https:\/\/h-in-q.com\/wp-content\/uploads\/2025\/12\/Backend-infrastructure-diagram-with-pipelines-640x427.webp 640w, https:\/\/h-in-q.com\/wp-content\/uploads\/2025\/12\/Backend-infrastructure-diagram-with-pipelines-1280x853.webp 1280w, https:\/\/h-in-q.com\/wp-content\/uploads\/2025\/12\/Backend-infrastructure-diagram-with-pipelines-768x512.webp 768w, https:\/\/h-in-q.com\/wp-content\/uploads\/2025\/12\/Backend-infrastructure-diagram-with-pipelines-320x213.webp 320w, https:\/\/h-in-q.com\/wp-content\/uploads\/2025\/12\/Backend-infrastructure-diagram-with-pipelines-1320x880.webp 1320w, https:\/\/h-in-q.com\/wp-content\/uploads\/2025\/12\/Backend-infrastructure-diagram-with-pipelines.webp 1536w\" sizes=\"auto, (max-width: 640px) 100vw, 640px\" \/><\/p>\n<ul>\n<li>Secure API gateways<\/li>\n<li>Scalable data pipelines<\/li>\n<li>Version control for models &amp; data<\/li>\n<li>Observability and logging<\/li>\n<li>Deployment automation<\/li>\n<\/ul>\n<p>&nbsp;<\/p>\n<h2>The 5% That Matters: Rethinking the \u201cAI Machine\u201d<\/h2>\n<h3>LLMs Are Engines, Not Whole Systems<\/h3>\n<p>Large Language Models (LLMs) are seductive because they seem intelligent. But the model itself is only <strong>a small part (about 5%) of what makes systems succeed<\/strong>. Without context, workflows, and guardrails, an LLM is an engine on a test bench; impressive but unusable. (<a href=\"https:\/\/www.mckinsey.com\/capabilities\/tech-and-ai\/our-insights\/charting-a-path-to-the-data-and-ai-driven-enterprise-of-2030?utm_source=chatgpt.com\" target=\"_blank\" rel=\"noopener\">McKinsey &amp; Company<\/a>)<\/p>\n<h3>Business Logic Lives Outside the Model<\/h3>\n<p>Business logic, compliance, exception handling, and integration with existing systems are the layers that let LLMs facilitate real decisions instead of isolated outputs.<\/p>\n<h3>AI Machine vs. Full AI System<\/h3>\n<table>\n<thead>\n<tr>\n<td><strong>Capability<\/strong><\/td>\n<td><strong>LLM Only<\/strong><\/td>\n<td><strong>Full AI System<\/strong><\/td>\n<\/tr>\n<\/thead>\n<tbody>\n<tr>\n<td>Language understanding<\/td>\n<td>\u2714<\/td>\n<td>\u2714<\/td>\n<\/tr>\n<tr>\n<td>Workflow execution<\/td>\n<td>\u2718<\/td>\n<td>\u2714<\/td>\n<\/tr>\n<tr>\n<td>Security &amp; Permission Policies<\/td>\n<td>\u2718<\/td>\n<td>\u2714<\/td>\n<\/tr>\n<tr>\n<td>Data Integration<\/td>\n<td>\u2718<\/td>\n<td>\u2714<\/td>\n<\/tr>\n<tr>\n<td>Compliance &amp; Auditing<\/td>\n<td>\u2718<\/td>\n<td>\u2714<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n<h2>Clean Data Is the Foundation, Not an Afterthought<\/h2>\n<h3>Data Quality Over Data Quantity<\/h3>\n<p>Many leaders fall into the \u201cAI imagine\u201d trap: assuming data size guarantees insight. In reality, <strong>clean, filtered, contextualized data is what powers reliable outcomes<\/strong>. Chaos in inputs leads to chaos in outputs, a lesson consistent with McKinsey findings that data excellence underpins transformation. (<a href=\"https:\/\/www.mckinsey.com\/capabilities\/tech-and-ai\/our-insights\/charting-a-path-to-the-data-and-ai-driven-enterprise-of-2030?utm_source=chatgpt.com\" target=\"_blank\" rel=\"noopener\">McKinsey &amp; Company<\/a>)<\/p>\n<h3>Contextualization Drives Value<\/h3>\n<p>Context gives meaning to information. A support transcript gains value only when tagged, linked to outcomes, and embedded in broader enterprise knowledge.<\/p>\n<h3>Enterprise Data Maturity for AI<\/h3>\n<p><img loading=\"lazy\" decoding=\"async\" class=\"aligncenter size-medium wp-image-6893\" src=\"https:\/\/h-in-q.com\/wp-content\/uploads\/2025\/12\/Enterprise-Data-Maturity-for-AI-640x427.jpg\" alt=\"Enterprise Data Maturity for AI\" width=\"640\" height=\"427\" srcset=\"https:\/\/h-in-q.com\/wp-content\/uploads\/2025\/12\/Enterprise-Data-Maturity-for-AI-640x427.jpg 640w, https:\/\/h-in-q.com\/wp-content\/uploads\/2025\/12\/Enterprise-Data-Maturity-for-AI-768x512.jpg 768w, https:\/\/h-in-q.com\/wp-content\/uploads\/2025\/12\/Enterprise-Data-Maturity-for-AI-320x213.jpg 320w, https:\/\/h-in-q.com\/wp-content\/uploads\/2025\/12\/Enterprise-Data-Maturity-for-AI.jpg 1092w\" sizes=\"auto, (max-width: 640px) 100vw, 640px\" \/><\/p>\n<h2>Architecting Intelligence: Platforms, Not Plugins<\/h2>\n<h3>Workflows Redesigned for Autonomy<\/h3>\n<p>AI cannot simply \u201cbolt onto\u201d existing business processes. To unlock autonomy, companies must redesign workflows so AI can interact at the right decision points, a principle highlighted in modern enterprise AI maturity frameworks. (<a href=\"https:\/\/www.mckinsey.com\/capabilities\/tech-and-ai\/our-insights\/charting-a-path-to-the-data-and-ai-driven-enterprise-of-2030?utm_source=chatgpt.com\" target=\"_blank\" rel=\"noopener\">McKinsey &amp; Company<\/a>)<\/p>\n<h3>Modular, Scalable Architecture Wins<\/h3>\n<p>A reliable platform must support:<\/p>\n<ul>\n<li>Scalable storage<\/li>\n<li>Pluggable compute<\/li>\n<li>Version tracking<\/li>\n<li>Automated retraining triggers<\/li>\n<\/ul>\n<h3>Platform vs. Plug\u2011In AI Models<\/h3>\n<table>\n<thead>\n<tr>\n<td><strong>Feature<\/strong><\/td>\n<td><strong>Plug\u2011In Tool<\/strong><\/td>\n<td><strong>Platform Architecture<\/strong><\/td>\n<\/tr>\n<\/thead>\n<tbody>\n<tr>\n<td>Scalability<\/td>\n<td>Low<\/td>\n<td>High<\/td>\n<\/tr>\n<tr>\n<td>Governance<\/td>\n<td>Weak<\/td>\n<td>Embedded<\/td>\n<\/tr>\n<tr>\n<td>Integration with Enterprise Systems<\/td>\n<td>Patchy<\/td>\n<td>Deep<\/td>\n<\/tr>\n<tr>\n<td>Monitoring &amp; Alerting<\/td>\n<td>Optional<\/td>\n<td>Standard<\/td>\n<\/tr>\n<tr>\n<td>Human\u2011Machine Interaction Control<\/td>\n<td>Basic<\/td>\n<td>Advanced<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n<p><img loading=\"lazy\" decoding=\"async\" class=\"aligncenter size-medium wp-image-6892\" src=\"https:\/\/h-in-q.com\/wp-content\/uploads\/2025\/12\/Modular-AI-platform-architecture-with-integration-data-and-governance-layers-640x427.webp\" alt=\"Modular AI platform architecture with integration, data, and governance layers\" width=\"640\" height=\"427\" srcset=\"https:\/\/h-in-q.com\/wp-content\/uploads\/2025\/12\/Modular-AI-platform-architecture-with-integration-data-and-governance-layers-640x427.webp 640w, https:\/\/h-in-q.com\/wp-content\/uploads\/2025\/12\/Modular-AI-platform-architecture-with-integration-data-and-governance-layers-1280x853.webp 1280w, https:\/\/h-in-q.com\/wp-content\/uploads\/2025\/12\/Modular-AI-platform-architecture-with-integration-data-and-governance-layers-768x512.webp 768w, https:\/\/h-in-q.com\/wp-content\/uploads\/2025\/12\/Modular-AI-platform-architecture-with-integration-data-and-governance-layers-320x213.webp 320w, https:\/\/h-in-q.com\/wp-content\/uploads\/2025\/12\/Modular-AI-platform-architecture-with-integration-data-and-governance-layers-1320x880.webp 1320w, https:\/\/h-in-q.com\/wp-content\/uploads\/2025\/12\/Modular-AI-platform-architecture-with-integration-data-and-governance-layers.webp 1536w\" sizes=\"auto, (max-width: 640px) 100vw, 640px\" \/><\/p>\n<h2>Governance and Security: The Non\u2011Negotiables<\/h2>\n<h3>No Guardrails, No Trust<\/h3>\n<p>AI systems without proper governance are risky. According to <strong>IBM\u2019s insights on AI data governance<\/strong>, ensuring the <strong>origin, sensitivity, and lifecycle of data<\/strong> is tracked and governed is essential before models are even deployed. (<a href=\"https:\/\/www.ibm.com\/think\/topics\/data-governance-for-ai?utm_source=chatgpt.com\" target=\"_blank\" rel=\"noopener\">ibm.com<\/a>) Security must be embedded at every layer.<\/p>\n<h3>Designing for Compliance and Accountability<\/h3>\n<p>Include mechanisms that:<\/p>\n<ul>\n<li>Track decisions and data lineage<\/li>\n<li>Enable human overrides<\/li>\n<li>Log outputs with context<\/li>\n<li>Restrict access based on roles<\/li>\n<\/ul>\n<h3>Governance Essentials<\/h3>\n<ul>\n<li>Permission tiers &amp; RBAC<\/li>\n<li>Immutable audit logs<\/li>\n<li>Model version control<\/li>\n<li>Regular compliance checks<\/li>\n<li>Ethical review boards<\/li>\n<\/ul>\n<h2>Adoption and Culture: Where AI Really Lives<\/h2>\n<h3>AI Isn\u2019t Adopted It\u2019s Absorbed<\/h3>\n<p>You can\u2019t deploy AI and walk away. Adoption depends on organizational culture, cross\u2011functional collaboration, and human\u2011in\u2011the\u2011loop design. Research frameworks for AI maturity (e.g., McKinsey) emphasize that adoption and change management are core pillars of enterprise success. (<a href=\"https:\/\/www.mckinsey.com\/capabilities\/tech-and-ai\/our-insights\/charting-a-path-to-the-data-and-ai-driven-enterprise-of-2030?utm_source=chatgpt.com\" target=\"_blank\" rel=\"noopener\">McKinsey &amp; Company<\/a>)<\/p>\n<h3>Human\u2011In\u2011The\u2011Loop Is Not Optional<\/h3>\n<p>Humans must be integrated for quality checks, validation, and continuous learning loops. Without tight coupling of human and machine roles, AI becomes a curiosity.<\/p>\n<h3>Adoption Best Practices<\/h3>\n<ul>\n<li>Internal AI champions<\/li>\n<li>Tailored training paths<\/li>\n<li>Transparent AI decision reporting<\/li>\n<li>Feedback loops for improvement<\/li>\n<\/ul>\n<h2>8. Stop Buying Tools Start Building Systems<\/h2>\n<h3>Tool Stack Culture Destroys Value<\/h3>\n<p>Measuring AI readiness by subscriptions purchased is dangerous. Tools are components, not architecture. Organizations that stack licenses yet ignore systemic thinking typically see sparse ROI and short\u2011lived initiatives.<\/p>\n<h3>System Thinking Is a Strategic Advantage<\/h3>\n<p>Leading firms like <strong>IBM\u2019s watsonx platform<\/strong> show what a holistic AI architecture can look like, combining data management, model governance, and deployment tools into one enterprise framework. (<a href=\"https:\/\/en.wikipedia.org\/wiki\/IBM_Watsonx?utm_source=chatgpt.com\" target=\"_blank\" rel=\"noopener\">Wikipedia<\/a>)<\/p>\n<h3>Strategic Shifts That Drive AI Success<\/h3>\n<ul>\n<li>Invest in platforms, not point tools<\/li>\n<li>Align KPIs to business outcomes<\/li>\n<li>Appoint product owners, not just project managers<\/li>\n<li>Build for evolution, not demos<\/li>\n<li>Treat AI like infrastructure, not an experiment<\/li>\n<\/ul>\n<p><img loading=\"lazy\" decoding=\"async\" class=\"aligncenter size-medium wp-image-6894\" src=\"https:\/\/h-in-q.com\/wp-content\/uploads\/2025\/12\/Strategic-AI-development-roadmap-integrated-into-business-planning-640x427.webp\" alt=\"Strategic AI development roadmap integrated into business planning\" width=\"640\" height=\"427\" srcset=\"https:\/\/h-in-q.com\/wp-content\/uploads\/2025\/12\/Strategic-AI-development-roadmap-integrated-into-business-planning-640x427.webp 640w, https:\/\/h-in-q.com\/wp-content\/uploads\/2025\/12\/Strategic-AI-development-roadmap-integrated-into-business-planning-1280x853.webp 1280w, https:\/\/h-in-q.com\/wp-content\/uploads\/2025\/12\/Strategic-AI-development-roadmap-integrated-into-business-planning-768x512.webp 768w, https:\/\/h-in-q.com\/wp-content\/uploads\/2025\/12\/Strategic-AI-development-roadmap-integrated-into-business-planning-320x213.webp 320w, https:\/\/h-in-q.com\/wp-content\/uploads\/2025\/12\/Strategic-AI-development-roadmap-integrated-into-business-planning-1320x880.webp 1320w, https:\/\/h-in-q.com\/wp-content\/uploads\/2025\/12\/Strategic-AI-development-roadmap-integrated-into-business-planning.webp 1536w\" sizes=\"auto, (max-width: 640px) 100vw, 640px\" \/><\/p>\n<p>&nbsp;<\/p>\n<h2>AI Is a System, Not a Shortcut<\/h2>\n<p>AI is among the most transformative technologies of our time, but only for those who build it right. The 95% of projects that fail do so not because the technology is flawed, but because leaders mistake AI for a tool instead of a <strong>comprehensive engineered system<\/strong>. Real AI success requires clean contextual data, platform\u2011first infrastructure, tight governance, redesigned workflows, and seamless human\u2011machine integration.<\/p>\n<p>Chatbot subscriptions and flashy demos are features. When enterprises stop chasing tools and start building platforms, AI becomes a strategic engine of value. The future belongs to those who engineer AI holistically not superficially.<\/p>\n<p>&nbsp;<\/p>\n<p><strong>References\u00a0<\/strong><\/p>\n<ol>\n<li><a href=\"https:\/\/sloanreview.mit.edu\/article\/why-so-many-high-profile-ai-projects-fail\/\" target=\"_blank\" rel=\"noopener\">Why So Many High-Profile AI Projects Fail (MIT Sloan)<\/a><\/li>\n<li><a href=\"https:\/\/www.mckinsey.com\/capabilities\/tech-and-ai\/our-insights\/charting-a-path-to-the-data-and-ai-driven-enterprise-of-2030\" target=\"_blank\" rel=\"noopener\">Charting a Path to the AI-Driven Enterprise (McKinsey)<\/a><\/li>\n<li><a href=\"https:\/\/www.ibm.com\/think\/topics\/data-governance-for-ai\" target=\"_blank\" rel=\"noopener\">Data Governance for AI: The New Frontier (IBM)<\/a><\/li>\n<li><a href=\"https:\/\/www2.deloitte.com\/us\/en\/pages\/consulting\/articles\/enterprise-ai-platform.html\" target=\"_blank\" rel=\"noopener\">Enterprise AI Adoption Maturity Report (Deloitte)<\/a><\/li>\n<li><a href=\"https:\/\/en.wikipedia.org\/wiki\/IBM_Watsonx\" target=\"_blank\" rel=\"noopener\">What Is IBM Watsonx? (Wikipedia)<\/a><\/li>\n<li><a href=\"https:\/\/h-in-q.com\/fr\/blog\/automating-data-pipelines-ai-impact\/\">Automating Data Pipelines for Real AI Impact<\/a><\/li>\n<\/ol>\n<p>&nbsp;<\/p>\n<p>&nbsp;<\/p>","protected":false},"excerpt":{"rendered":"<p>The AI Illusion That\u2019s Crippling Modern Business We are living through one of the most dangerous illusions in corporate technology today: the belief that a chatbot subscription equals an AI solution. Too many executives think that \u201cAI imagine\u201d means buying ChatGPT and layering a conversational interface on a simple use case. According to research tied&#8230;<\/p>","protected":false},"author":10,"featured_media":6895,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"_jetpack_newsletter_access":"","_jetpack_dont_email_post_to_subs":false,"_jetpack_newsletter_tier_id":0,"_jetpack_memberships_contains_paywalled_content":false,"_jetpack_feature_clip_id":0,"_jetpack_memberships_contains_paid_content":false,"footnotes":"","jetpack_post_was_ever_published":false},"categories":[56],"tags":[],"class_list":["post-6889","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-ai-solutions"],"jetpack_sharing_enabled":true,"jetpack_featured_media_url":"https:\/\/h-in-q.com\/wp-content\/uploads\/2025\/12\/AI-Is-Not-a-Chatbot-The-Real-Reason-95-of-AI-Projects-Fail.webp","_links":{"self":[{"href":"https:\/\/h-in-q.com\/fr\/wp-json\/wp\/v2\/posts\/6889","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/h-in-q.com\/fr\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/h-in-q.com\/fr\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/h-in-q.com\/fr\/wp-json\/wp\/v2\/users\/10"}],"replies":[{"embeddable":true,"href":"https:\/\/h-in-q.com\/fr\/wp-json\/wp\/v2\/comments?post=6889"}],"version-history":[{"count":2,"href":"https:\/\/h-in-q.com\/fr\/wp-json\/wp\/v2\/posts\/6889\/revisions"}],"predecessor-version":[{"id":7095,"href":"https:\/\/h-in-q.com\/fr\/wp-json\/wp\/v2\/posts\/6889\/revisions\/7095"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/h-in-q.com\/fr\/wp-json\/wp\/v2\/media\/6895"}],"wp:attachment":[{"href":"https:\/\/h-in-q.com\/fr\/wp-json\/wp\/v2\/media?parent=6889"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/h-in-q.com\/fr\/wp-json\/wp\/v2\/categories?post=6889"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/h-in-q.com\/fr\/wp-json\/wp\/v2\/tags?post=6889"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}