B-TIER technique
Fully synthetic AI creator — Higgsfield, ElevenLabs, Veo and CapCut stack
100k ft — What is this?
Fully synthetic AI creator — Higgsfield, ElevenLabs, Veo and CapCut stack
50k ft — Why care?
How viral fake-influencer videos are made: generate the character in Higgsfield or Midjourney, voice it with ElevenLabs, animate with an AI video model like Veo or Runway, and stitch in CapCut. Useful as a pipeline map and as a detection primer.
10k ft — Key takeaways
• Character image, voice, animation and edit are four separable stages.
• Viral footage can now be fully synthetic; verify sources.
5k ft — Tools & techniques
CapCut, ElevenLabs, Higgsfield, Midjourney, Runway, Stitch, Veo
AI character generation, image-to-video, voice synthesis, workflow automation
1k ft — Deep-dive / try it
Comment 'FAKE' and I'll send it over.
Transcript excerpt
So this video got millions of views and none of it is real person AI even the footage AI Here's how people are making these first generate the character in a tool like Higgs field or mid-journey Then use 11 laps for the voice after that animate everything with an AI video model like VO or runway and you stitch it All together in CapCut and now you've got a fully fake creator that looks real enough to go viral So just comment fake and I'll send you the whole workflow
More from @nateherkai
- A-TIER Claude Opus 5.5 to Blender — 3D animation built entirely as code
- A-TIER Claudex skill — Claude and GPT-6 Astra cross-audit plan and build
- B-TIER Google game Playground — why a harness that can't see its output fails
- A-TIER Claude trading watcher — LiquidTrade + QuiverQuant for insider-trade alerts
- A-TIER Claude design stack — Emil Kowalski, Impeccable, taste skills plus Figma MCP