The 3-Step Realistic AI Ad Workflow: Locking the stadium, street & office

2026-08-10 · Academy lesson · Adil · recording, 1:32 · original

Topic: Assets, references & consistency · mode: Tutorial · tags: course chapter, series, one creator's experience

In the locations lesson of The 3-Step Realistic AI Ad Workflow, Adil has Claude write the stadium prompt from three cues (running track, three-quarter view, sunny day), tests two stadiums in video with the same character sheet and one line, keeps the brighter morning one, and locks the street corner and office the same way.

Part of another source

This recording is almost entirely part of 3-Step Workflow To Make Ultra-Realistic AI Ads (2026-06-23); what is said in it is covered on that page.

What is said

• 00:00:00 — The stadium, street corner and office: Claude-written prompts, testing options on the canvas, locking the winners

Threads in this video

• restate · Which LLM should be used to write video prompts? — Rather than handwriting a prompt for a cinematic stadium, Adil gives Claude three cues (running track, three-quarter view, sunny day), and Claude returns a full prompt describing the track, the field and the light. (00:00:00)
• restate · How do you keep a location, product or prop the same across shots? — As with the kitchen of his ad, Adil generates each location at a three-quarter angle, which he says gives a better hit rate. (00:00:00)
• restate · How do you spend fewer credits? — Adil says testing every location in video only feels slow: it avoids wasting many credits later on full scenes generated with the wrong assets. (00:01:22)

Mentioned

Canvas · character sheet · Claude

Only in the description or credits: Higgsfield Academy · The 3-Step Realistic AI Ad Workflow

← Previous: Build a Brand's Visuals with AI: Marketing Studio Try-Ons · Next → Build an Ultra-Realistic Short Film in 4K: Locations with the 3/4 angle