{"id":36009,"date":"2026-08-27T05:03:00","date_gmt":"2026-08-27T05:03:00","guid":{"rendered":"https:\/\/www.sozee.ai\/resources\/use-ai-creative-director-content\/"},"modified":"2026-09-02T14:28:09","modified_gmt":"2026-09-02T14:28:09","slug":"use-ai-creative-director-content","status":"publish","type":"post","link":"https:\/\/www.sozee.ai\/resources\/use-ai-creative-director-content\/","title":{"rendered":"How to Use an AI Creative Director: 7-Step Revenue System"},"content":{"rendered":"<h2 id=\"key-takeaways\">Key Takeaways<\/h2>\n<ul>\n<li>Traditional AI prompting is inconsistent, so treat AI as a repeatable studio system by building a Brand Brain once and directing every frame.<\/li>\n<li>Define machine-readable voice, visual, and monetization rules in a single Brand Brain document to eliminate generic output and voice drift.<\/li>\n<li>Generate concepts before assets, then apply a 1\u20135 revenue-tied rubric that blocks any idea scoring below 4 on accuracy, voice, engagement, or revenue alignment.<\/li>\n<li>Convert one approved concept into a full platform ecosystem matrix, such as Instagram carousels, TikTok reels, and SFW-to-NSFW arcs, using locked likeness and saved environments for consistency.<\/li>\n<li>Sozee turns this seven-step workflow into an operational studio: <a href=\"https:\/\/app.sozee.ai\/sign-up\" target=\"_blank\"><strong>start building your repeatable content system<\/strong><\/a> to replace slot-machine prompting with predictable results.<\/li>\n<\/ul>\n<h2>Step 1: Build Your Single-Source Brand Brain<\/h2>\n<p>Every repeatable content system begins with a single source of truth. <a href=\"https:\/\/alstudio.ai\/Blog\/blog-brand-dna\" target=\"_blank\" rel=\"noindex nofollow\">Brand DNA operates at three layers<\/a>, which are strategic (purpose, mission, values, positioning), expressive (personality, voice, tone), and sensory (visual identity, imagery, motion). Together these layers enable instant recognition before a logo appears. These three layers form the foundation of brand consistency, but AI cannot execute on abstract ideas like \u201capproachable\u201d or \u201cwarm\u201d without clear rules. The Brand Brain translates all three layers into machine-readable rules that AI can apply consistently across every asset.<\/p>\n<p>The Brand Brain is the machine-readable version of all three layers, translated into rules an AI can execute. To build your Brand Brain, define the following four categories and store them in one document, because each category maps to the strategic, expressive, and sensory layers in a format AI can follow.<\/p>\n<figure style=\"text-align: center;\"><a href=\"https:\/\/app.sozee.ai\/sign-up\" target=\"_blank\"><img src=\"https:\/\/cdn.aigrowthmarketer.co\/1759125608311-5672a1d609fd.png\" alt=\"Use the Curated Prompt Library to generate batches of hyper-realistic content.\" style=\"max-height: 500px;\" loading=\"lazy\" decoding=\"async\"><\/a><figcaption><em>Use the Curated Prompt Library to generate batches of hyper-realistic content.<\/em><\/figcaption><\/figure>\n<ul>\n<li><strong>Voice pillars:<\/strong> Replace adjectives with concrete rules. <a href=\"https:\/\/sameness.co\/ai-brand-governance\" target=\"_blank\" rel=\"noindex nofollow\">Machine-readable constraints<\/a> look like <em>max_sentence_length: 18 words<\/em>, <em>contractions: required<\/em>, and <em>banned_words: [leverage, synergy, seamless]<\/em>, not \u201cconfident but approachable.\u201d<\/li>\n<li><strong>Visual rules:<\/strong> Specify exact hex values, lighting direction, composition rules, and film stock references. <a href=\"https:\/\/memorable-studio.com\/blog\/ai-brand-consistency\" target=\"_blank\" rel=\"noindex nofollow\">Memorable Studio&#8217;s Brand Visual DNA framework<\/a> converts abstract descriptions like \u201cwarm and welcoming\u201d into measurable parameters such as colour temperature 5200K and saturation \u221215%.<\/li>\n<li><strong>Monetization KPIs:<\/strong> Document target engagement rate (aim for &gt;4%), revenue per post type, and platform priority order.<\/li>\n<li><strong>Reusable asset libraries:<\/strong> Create saved environments, outfit combinations, and prop sets that attach via Sozee&#8217;s @-reference system without leaving the prompt bar.<\/li>\n<\/ul>\n<blockquote>\n<p><strong>\u26a0 Common Pitfall:<\/strong> Generic output from vague adjectives. \u201cPlayful and bold\u201d gives an AI nothing it can act on.<\/p>\n<p><strong>\u2705 Pro Tip:<\/strong> Convert every brand rule into a machine-readable constraint. <a href=\"https:\/\/glean.com\/perspectives\/why-you-need-both-brand-guidelines-and-ai-prompts-for-consistency\" target=\"_blank\" rel=\"noindex nofollow\">Log every human editorial correction as a new Brand Brain rule<\/a>. If you rewrote a caption three times, that rewrite is a constraint waiting to be encoded.<\/p>\n<h2>Step 2: Assign the AI Creative Director Role<\/h2>\n<p>A prompt is a wish, a one-time request with no memory. A director role is a decision framework that persists across sessions. To establish that framework, feed the Brand Brain as persistent context at the start of every session. <a href=\"https:\/\/sameness.co\/ai-brand-governance\" target=\"_blank\" rel=\"noindex nofollow\">Without persistent brand context delivered at session start, AI agents lose institutional knowledge between tasks and default to generic training data<\/a>, which causes cumulative voice drift.<\/p>\n<p>Use this 10-element prompt template to assign the role:<\/p>\n<ol>\n<li>You are a creative director for [Brand Name].<\/li>\n<li>Your audience is [persona description].<\/li>\n<li>Voice rules: [paste machine-readable constraints from Brand Brain].<\/li>\n<li>Visual rules: [paste sensory layer parameters].<\/li>\n<li>Monetization goal: [specific KPI, for example drive Fanvue subscriptions].<\/li>\n<li>Platform priority: [for example TikTok first, Instagram second].<\/li>\n<li>Banned vocabulary: [list].<\/li>\n<li>Approved asset library: [list saved environments, outfits, objects by @-reference name].<\/li>\n<li>Self-critique requirement: Score every concept 1\u20135 across accuracy, brand voice, engagement potential, and revenue alignment before presenting it.<\/li>\n<li>Only present concepts that score 4 or above on all four dimensions.<\/li>\n<\/ol>\n<p>In Sozee, this context lives in the Agent. The Agent reads your characters, your library, and your performance data, then proposes and produces. It answers with working controls, not paragraphs.<\/p>\n<figure style=\"text-align: center;\"><a href=\"https:\/\/app.sozee.ai\/sign-up\" target=\"_blank\"><img src=\"https:\/\/cdn.aigrowthmarketer.co\/1762997925636-7453a7a8b2ad.png\" alt=\"Sozee AI Platform\" style=\"max-height: 500px;\" loading=\"lazy\" decoding=\"async\"><\/a><figcaption><em>Sozee AI Platform<\/em><\/figcaption><\/figure>\n<h2>Step 3: Generate Big Ideas Before Assets<\/h2>\n<p>Once the director role is assigned and the Brand Brain is loaded, the next decision is what to create. The most common workflow error is jumping straight to image generation without defining the strategic concept first. <a href=\"https:\/\/kompozy.io\/guides\/ai-creative-director-workflows\" target=\"_blank\" rel=\"noindex nofollow\">Humans retain ownership of the strategic concept, reading cultural moments and setting the angle, while AI handles the format matrix and production<\/a>. Run the director prompt to produce three to five concepts first. Each concept should include a core insight, a platform-native hook, and a monetization angle. No images are generated at this stage. Concepts are text. Assets come later.<\/p>\n<p>If you want a faster starting point, Sozee&#8217;s Explore feed accelerates this step by surfacing ready-made concepts, such as social, selfie, fitness, and outdoors, that generate instantly with your locked character. One click produces a platform-aware concept, which can replace the manual concept-generation phase for routine content.<\/p>\n<h2>Step 4: Run a Brutal Self-Critique Loop Tied to Revenue<\/h2>\n<p>Apply a 1\u20135 rubric to every concept before a single asset is produced. <a href=\"https:\/\/bot365.co.uk\/ai-output-evaluation-rubric-for-marketing-teams-accuracy-brand-voice-and-risk\" target=\"_blank\" rel=\"noindex nofollow\">Marketing teams can use a reusable scoring rubric across accuracy, brand voice, and risk dimensions, with automatic blocking of any output that receives a red-flag score in any dimension regardless of the average<\/a>.<\/p>\n<p>Score each concept across these four dimensions, which together keep concepts brand-safe and revenue-effective:<\/p>\n<ul>\n<li><strong>Accuracy (1\u20135):<\/strong> Does the concept reflect verified, brand-safe claims?<\/li>\n<li><strong>Brand voice (1\u20135):<\/strong> Does it match the machine-readable voice constraints in the Brand Brain?<\/li>\n<li><strong>Engagement potential (1\u20135):<\/strong> Does it contain a platform-native hook, a reason to share, and a completion driver?<\/li>\n<li><strong>Revenue alignment (1\u20135):<\/strong> Does it drive the monetization KPI defined in Step 1?<\/li>\n<\/ul>\n<p>Any concept scoring below 4 on any single dimension is automatically blocked, not averaged up. <a href=\"https:\/\/nailedit.ai\/prompts\/analytics-insights\" target=\"_blank\" rel=\"noindex nofollow\">Content generating high traffic but zero conversions is lower priority than lower-traffic content that produces measurable leads or revenue<\/a>, so weak concepts do not move forward even if they seem creative.<\/p>\n<blockquote>\n<p><strong>\u26a0 Common Pitfall:<\/strong> Voice loss. High-volume AI output regresses toward a recognizable sameness, hedged, over-explained, and stock-lit, that audiences learn to tune out.<\/p>\n<p><strong>\u2705 Pro Tip:<\/strong> Apply the same feedback loop here. When you edit a concept during the rubric review, that edit becomes a new constraint that prevents the same issue in future concepts.<\/p>\n<h2>Step 5: Turn One Approved Concept Into a Platform Ecosystem Matrix<\/h2>\n<p>Once a concept clears the rubric, one approved concept becomes a full platform ecosystem. <a href=\"https:\/\/conbersa.ai\/learn\/what-is-platform-native-content\" target=\"_blank\" rel=\"noindex nofollow\">Brands that tailor content for each platform<\/a> often achieve higher engagement rates than those that cross-post identical content, and TikTok explicitly suppresses content carrying competing platform watermarks.<\/p>\n<p>Expand the concept into the following asset types using Sozee&#8217;s Photo Shoot and video tools:<\/p>\n<figure style=\"text-align: center;\"><a href=\"https:\/\/app.sozee.ai\/sign-up\" target=\"_blank\"><img src=\"https:\/\/sozee.ai\/wp-content\/uploads\/2025\/11\/Sozee-60-Seconds-To-Generate-Content-White.gif\" alt=\"GIF of Sozee Platform Generating Images Based On Inputs From Creator on a White Background\" style=\"max-height: 500px;\" loading=\"lazy\" decoding=\"async\"><\/a><figcaption><em>GIF of Sozee Platform Generating Images Based On Inputs From Creator on a White Background<\/em><\/figcaption><\/figure>\n<ul>\n<li><strong>Instagram static and carousel:<\/strong> Locked likeness, saved environment, 4:5 ratio, and a brand-voice caption.<\/li>\n<li><strong>TikTok reel:<\/strong> 9:16 vertical, native hook in the first second, and Reel Cloning to rebuild proven motion formats in your likeness.<\/li>\n<li><strong>Story arc:<\/strong> Three-frame sequence using the same outfit and environment, with an SFW teaser leading to a monetized destination.<\/li>\n<li><strong>SFW-to-NSFW arc (where applicable):<\/strong> Photo Shoot builds a locked, coherent set of up to ten images with the ramp and the ceiling set by you, not the algorithm.<\/li>\n<\/ul>\n<blockquote>\n<p><strong>\u2705 Pro Tip:<\/strong> Pre-build environments and set SFW and NSFW ceilings before the shoot begins. In Sozee, a saved environment built from up to four reference photos stays consistent across every image in the set, so you build the bedroom once and shoot in it for a year. Set the content ceiling per platform in advance so every asset in the batch is compliant before it reaches the Vault.<\/p>\n<h2>Step 6: Add the Human Approval Gate and Schedule Content<\/h2>\n<p><a href=\"https:\/\/kompozy.io\/guides\/ai-creative-director-workflows\" target=\"_blank\" rel=\"noindex nofollow\">A human review gate over finished assets allows a director to clear a week of content in the time previously required to brief a single piece<\/a>. The 70% time saving appears here, not by removing humans, but by positioning them at the decision point rather than the production point.<\/p>\n<p>The approval gate checks three things:<\/p>\n<ul>\n<li>Likeness consistency: identical character rendering across every frame.<\/li>\n<li>Brand voice: caption matches machine-readable constraints from the Brand Brain.<\/li>\n<li>Platform compliance: aspect ratio, caption length, and content ceiling are correct per platform.<\/li>\n<\/ul>\n<p>Approved assets move directly into Sozee&#8217;s Scheduler. Connect Instagram, TikTok, X, Facebook, Reddit, and Fanvue per character, not per account. Set per-platform captions, preview the live post, and schedule. The &gt;4% engagement target is maintained because every asset was platform-native before it was scheduled, not reformatted after the fact.<\/p>\n<p><a href=\"https:\/\/app.sozee.ai\/sign-up\" target=\"_blank\"><strong>Start creating now\u2014build your first platform ecosystem in under an hour.<\/strong><\/a><\/p>\n<h2>Step 7: Feed Performance Data Back Into the Brand Brain<\/h2>\n<p>The system compounds only when performance data flows back into the Brand Brain. Sozee&#8217;s Analytics splits what Sozee posted from what you posted manually, so the contribution of the workflow is measurable in isolation. <a href=\"https:\/\/tareno.co\/resources\/blog\/ai-assisted-content-workflow\" target=\"_blank\" rel=\"noindex nofollow\">Common failure modes in AI content workflows include failing to connect post-publish reporting back to new brief creation<\/a>, which stalls improvement.<\/p>\n<p>After each publishing cycle, update the Brand Brain with:<\/p>\n<ul>\n<li>Top-performing hooks, which you add to the approved prompt library.<\/li>\n<li>Underperforming formats, which you convert into suppression rules in the rubric.<\/li>\n<li>Any human edit made during the approval gate, which you encode as a new constraint.<\/li>\n<li>Engagement and revenue data per asset type, which you use to update monetization KPI weighting.<\/li>\n<\/ul>\n<p><a href=\"https:\/\/bot365.co.uk\/ai-output-evaluation-rubric-for-marketing-teams-accuracy-brand-voice-and-risk\" target=\"_blank\" rel=\"noindex nofollow\">Teams should calibrate the rubric quarterly by rescoring 20\u201330 recent AI-assisted outputs, identifying recurring weaknesses, then updating prompt templates and approval thresholds accordingly<\/a>.<\/p>\n<h2>Success Metrics That Prove the System Works<\/h2>\n<p>Three benchmarks confirm that the workflow is operating correctly and that each step is pulling its weight.<\/p>\n<ul>\n<li><strong>Doubled weekly output:<\/strong> A solo creator using a structured AI content workflow with batching can often produce more publication-ready posts per week at a similar quality level compared to without AI. Sozee&#8217;s Photo Shoot turns one approved concept into a full locked set, which multiplies that baseline further.<\/li>\n<li><strong>Greater than 4% engagement:<\/strong> The &gt;4% engagement target is achievable because the ecosystem matrix in Step 5 ensures every asset is platform-native before scheduling, not reformatted after the fact.<\/li>\n<li><strong>70% less production time:<\/strong> <a href=\"https:\/\/memorable-studio.com\/blog\/ai-brand-consistency\" target=\"_blank\" rel=\"noindex nofollow\">Implementing a Brand Visual DNA and structured prompt architecture reduces prompt iterations per image from 15\u201330 to 3\u20136 and campaign time for 30 assets from 3\u20135 weeks to 10\u201315 business days<\/a>.<\/li>\n<\/ul>\n<h2>Advanced Next Steps to Scale Your Studio<\/h2>\n<p>Once the seven-step workflow is running, three extensions increase output and revenue ceiling without adding production hours.<\/p>\n<p><strong>Agent-assisted batching.<\/strong> Sozee&#8217;s Agent takes a half-formed idea, interviews you into a finished setup, and writes straight into the prompt bar and Photo Control panel. When the conversation ends, the shoot sits one tap from Generate. Use it to batch an entire week of content in a single session before you move to more advanced scaling.<\/p>\n<p><strong>Multi-character workspaces.<\/strong> Agencies and creators managing multiple personas run each character in an isolated workspace with its own Vault, connected accounts, and credits, so one login supports every client while keeping them fully separated. AI-native workflows can enable higher levels of output per producer when voice profiles and multi-agent review gates are enforced across these workspaces.<\/p>\n<p><strong>Quarterly Brand Brain audits.<\/strong> <a href=\"https:\/\/netcentric.biz\/insights\/2026\/05\/brand-governance-in-ai-era\" target=\"_blank\" rel=\"noindex nofollow\">Brand governance in the AI era requires governing the encoded brand as versioned living infrastructure<\/a>, not a static PDF reviewed once a year. Schedule a quarterly audit to rescore recent outputs, update machine-readable constraints, and retire underperforming asset templates so the system keeps improving.<\/p>\n<h2>Frequently Asked Questions<\/h2>\n<h3>How do I keep likeness consistent across every platform?<\/h3>\n<p>Likeness consistency is a setup decision, not a generation decision. In Sozee, you upload three reference photos and the system reconstructs your likeness with locked parameters that persist across every image and video generated in that character&#8217;s workspace. Photo Control&#8217;s five dimensions, which are Setting, Outfit, Shot style, Expression, and Object, let you change everything around the character without touching the character itself. Your locked character appears consistently whether you are generating a TikTok reel, an Instagram carousel, or a Fanvue set. Saved environments built from up to four reference photos ensure the room stays the room across every shoot. Likeness drift is a prompt problem, and Sozee replaces the prompt with a director&#8217;s panel where the character is a fixed constant.<\/p>\n<h3>What ROI can I expect from an AI creative director workflow?<\/h3>\n<p>ROI shows up across three dimensions, which are output volume, time recovered, and revenue per asset. Structured AI content workflows typically deliver 3\u20135x content velocity gains, measured as asset volume or cycle-time reductions, in <a href=\"https:\/\/www.thestarrconspiracy.com\/insights\/benchmarks\/ai-content-workflow-benchmarks-2025\" target=\"_blank\" rel=\"noindex nofollow\">benchmarks that incorporate governance metrics<\/a>. On the time side, automation of manual formatting, reformatting, and distribution tasks can save substantial hours per week for a typical content team. On the revenue side, influencer marketing ROI averages $5.78 for every $1 spent across the broader creator economy. For individual creators, the more direct gain is the ability to take on more brand deals, post more frequently to subscription platforms, and stop turning down opportunities because production hours ran out. The workflow described here is designed to deliver doubled weekly output, greater than 4% engagement, and 70% less production time, which you can measure against your baseline within the first four weeks.<\/p>\n<h3>How much human oversight is still required?<\/h3>\n<p>Human oversight concentrates at two points, which are concept selection in Step 3 and the approval gate in Step 6. AI handles ideation volume, self-critique scoring, asset production, platform formatting, and scheduling. Humans decide which concept moves forward and whether the finished assets clear the brand voice and likeness check before publishing. In practice, the approval gate for a week of content takes minutes rather than hours because every asset arrives pre-formatted, pre-captioned, and pre-scored. The division of labor stays stable, with AI excelling at production and variation while humans retain the taste calls and cultural reads that define a creative director&#8217;s judgment. The Brand Brain encodes those taste calls as machine-readable constraints so the AI applies them at generation time, which reduces the volume of corrections needed at the approval gate over successive cycles.<\/p>\n<h3>Can the system handle both SFW and monetized NSFW content safely?<\/h3>\n<p>Yes. Sozee&#8217;s Photo Shoot feature builds a full SFW-to-NSFW arc within a single locked set, with the ramp and the ceiling set by the creator before generation begins. This means the content ceiling is a deliberate decision encoded into the shoot setup, not a post-production filter applied inconsistently. SFW teasers for Instagram and TikTok and monetized sets for Fanvue are generated from the same locked character and environment in the same session, then routed to the correct platform via the Scheduler. Compliance and verification are built into the character setup process, not bolted on afterward. The Brand Brain stores per-platform content ceilings as constraints, so the ecosystem matrix in Step 5 automatically applies the correct ceiling to each asset type without manual review of every individual image.<\/p>\n<h2>Ready to Turn One Idea Into a Full Content Studio?<\/h2>\n<p>The seven-step AI creative director workflow replaces slot-machine prompting with a repeatable studio system. Build the Brand Brain once. Assign the director role. Generate concepts before assets. Score every concept against revenue metrics. Expand one approved idea into a full platform ecosystem with locked likeness. Clear the approval gate in minutes. Feed performance data back into the system. Repeat.<\/p>\n<p>The result is doubled weekly output, greater than 4% engagement, and 70% less time in production, without sacrificing brand voice, visual consistency, or the monetization potential that makes content worth creating in the first place.<\/p>\n<p>Sozee is the execution layer that makes every step of this system operational. Photo Control, Photo Shoot, Agent, Vault, and Scheduler work together so that direction replaces prompting and one idea becomes a full content studio.<\/p>\n<p><a href=\"https:\/\/app.sozee.ai\/sign-up\" target=\"_blank\"><strong>Build your content system\u2014sign up for Sozee and run your first AI creative director workflow in under an hour.<\/strong><\/a><\/p>\n<section data-read-next=\"true\">\n<h2>Read Next<\/h2>\n<ul>\n<li><a href=\"https:\/\/sozee.ai\/resources\/consistent-revenue-automated-ai-content\" target=\"_blank\">How to Build Consistent Revenue with Automated AI Content<\/a><\/li>\n<li><a href=\"https:\/\/sozee.ai\/resources\/ai-content-boost-creator-revenue\" target=\"_blank\">How to Use AI Content to Boost Creator Revenue by 43%<\/a><\/li>\n<li><a href=\"https:\/\/sozee.ai\/resources\/scale-content-for-creator-agencies\" target=\"_blank\">Scale Content for Creator Agencies: 7 AI Strategies to 10x<\/a><\/li>\n<li><a href=\"https:\/\/sozee.ai\/resources\/creator-content-production-process\" target=\"_blank\">Creator Content Production Process: 7-Step AI Workflow<\/a><\/li>\n<li><a href=\"https:\/\/sozee.ai\/resources\/integrate-ai-into-content-strategy\" target=\"_blank\">How to Integrate AI Into Your Content Strategy: 7-Step Guide<\/a><\/li>\n<\/ul>\n<\/section>\n","protected":false},"excerpt":{"rendered":"<p>Turn one idea into a full content studio. Sozee&#8217;s 7-step AI creative director system builds brand consistency and drives real revenue. Start today.<\/p>\n","protected":false},"author":2,"featured_media":36008,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[8,10],"tags":[],"class_list":["post-36009","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-automation","category-monetization"],"_links":{"self":[{"href":"https:\/\/www.sozee.ai\/resources\/wp-json\/wp\/v2\/posts\/36009","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.sozee.ai\/resources\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.sozee.ai\/resources\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.sozee.ai\/resources\/wp-json\/wp\/v2\/users\/2"}],"replies":[{"embeddable":true,"href":"https:\/\/www.sozee.ai\/resources\/wp-json\/wp\/v2\/comments?post=36009"}],"version-history":[{"count":1,"href":"https:\/\/www.sozee.ai\/resources\/wp-json\/wp\/v2\/posts\/36009\/revisions"}],"predecessor-version":[{"id":43382,"href":"https:\/\/www.sozee.ai\/resources\/wp-json\/wp\/v2\/posts\/36009\/revisions\/43382"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.sozee.ai\/resources\/wp-json\/wp\/v2\/media\/36008"}],"wp:attachment":[{"href":"https:\/\/www.sozee.ai\/resources\/wp-json\/wp\/v2\/media?parent=36009"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.sozee.ai\/resources\/wp-json\/wp\/v2\/categories?post=36009"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.sozee.ai\/resources\/wp-json\/wp\/v2\/tags?post=36009"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}