The task
You have an image or video and 5 minutes to ship a caption. The risk is writing one caption and copying it to every platform; captions that are native on X die on LinkedIn, and Xiaohongshu has its own grammar. The job is to use AI for 3 variants per platform, each adapted to that platform’s reader and CTA convention, in a tone that does not give itself away as AI.
When AI helps — and when it does not
AI is great at varying tone (punchy / curious / contrarian) and switching vocabulary register per platform. It is bad at current native idiom: what’s cringe on TikTok this month may have been fine last quarter. Feed AI 2-3 recent captions you admire as reference; otherwise it defaults to a stale “creator” tone.
What to feed the AI
- Asset description (image / video / carousel: what is visible, what is happening)
- Platform list with target reader per platform
- Brand voice and banned tics (no ”🎯,” no “Let’s dive in”)
- CTA per platform: different platforms reward different asks
- Recent reference captions you admire (one per platform)
- Hashtag policy: Instagram up to 10, X 1-2, LinkedIn 3, TikTok niche-only
Copy-ready prompt
Write 3 caption variants per platform.
Asset: <description of what's in the image / video>
Brand voice: <one sentence>
Banned tics: <list, including "🎯", "Let's dive in", emoji bullets>
For each platform, here's the reader and a reference caption I admire:
X (Twitter)
Reader: <line>
CTA: <reply / link / no CTA>
Reference: "<paste>"
Hashtag policy: 1-2 max
LinkedIn
Reader: <line>
CTA: <comment / DM / link>
Reference: "<paste>"
Hashtag policy: 3 niche
Instagram
Reader: <line>
CTA: <save / share / link in bio>
Reference: "<paste>"
Hashtag policy: 5-10, mix of broad and niche
Xiaohongshu
Reader: <line>
CTA: <collect / follow>
Reference: "<paste>"
Hashtag policy: 3-5 niche
Return per platform: 3 variants in these tones — punchy / curious / contrarian. For each variant, include: caption text, hashtags, and one sentence on which type of reader is most likely to engage.
Do not use the same opening word across variants. Vary length within the platform's norm.For repurposing one asset: “Now produce a 12-second voiceover script that the same asset could carry on TikTok / Shorts / Reels.”
Recommended output structure
Per platform: 3 variants in a small table (tone / caption / hashtags / target reader). A short note at the bottom on which platform you should ship first based on what is in the asset.
How to check the output is usable
- Each variant matches the platform’s reference vocabulary
- Hashtag counts respect the platform’s norm
- CTAs use the platform’s native verb (Reply on X, Comment on LinkedIn, Save on IG)
- Opening words differ across variants (no “AI cadence”)
- A stranger from your target audience would re-share the strongest variant
Common mistakes
- One caption across all platforms: performs poorly everywhere
- LinkedIn voice on X (over-formal) or X voice on LinkedIn (too curt)
- Hashtag overstuffing. Instagram’s 30-cap habit still exists; 5-10 is the new norm
- Letting AI repeat its favourite emoji: they age fast
- Generic CTAs (“link in bio”) on platforms that punish them
- Captions that summarise the asset; captions should add, not narrate
Practical depth notes
For AI Social Captions: Platform Tone for X, LinkedIn, IG, Xiaohongshu, the difference between a usable AI result and a generic one is the input packet. Give the model the audience, the current draft or raw material, the desired format, the decision you need to make, and two examples of what good and bad output look like. Ask it to preserve facts first, then improve structure or wording second.
After the first response, do a separate review pass. Look for missing constraints, invented details, weak calls to action, and language that sounds plausible but does not match the real situation. The best final output should be easy to use immediately: clear owner, clear next step, and no hidden assumption that someone else has to untangle. A stronger version of this workflow also defines the handoff. Decide who will use the output, what they should do next, and what information would make them reject it. If the deliverable is copy, test whether it has a single clear action. If it is analysis, test whether it separates observation from recommendation. If it is planning, test whether dates, owners, and tradeoffs are explicit enough for someone else to execute.
FAQ
- Should I A/B captions on the same asset? Yes, by reposting at different times. Each platform’s algorithm penalises identical re-posts within 24h.
- What about alt text? Always write it: accessibility, but also platforms parse it.
- Should AI write the hashtags? Yes, but verify niche hashtags exist and are active. AI invents.
Related
- X thread AI: when the caption becomes a thread
- Cross-platform repurpose: multi-platform reuse
- Content calendar: when captions ship
- TikTok caption: TikTok-specific patterns
- Reel hook: when the asset is a Reel and needs a hook
- Reel caption prompts: additional Reel caption phrasings
- TikTok caption prompts: additional TikTok caption phrasings