Creative director runs one-week AI vs human video challenge
A creative director with 150+ software videos directed is spending a week making the same script three ways: AI solo with Opus 5.5, AI plus human work using ChatGPT images animated in Higgsfield, and a human team, then comparing the results.
I'm starting a 1-week challenge: AI vs Human Day 0. Over the last 2 years, my studio has made 220+ videos for 85+ software brands. I've personally curated/directed at least 150 of them. Now I have 1 week to find out: 1. If I can produce a studio-level video MYSELF alone, without a team of designer/animators to help me, using only my judgement and AI tools. 2. If AI can produce anything decent on its own (what an average person can expect, if he's not a creative person with lots of video experience). 3. How many artifacts and bugs I'll face on the way. And whether these bugs ruin the video's value due to the common perception "if AI generated = worthless". 4. And then we'll compare it with a video made by a team of humans (Astra Motion). So, overall, 3 approaches: 1/ AI solo ↳ I give it script only, then use prompts to refine. Not touching anything manually. Opus 5.5 is my choice here. 2/ AI + human ↳ Script is the same. I use ChatGPT to generate images (storyboard), then "animate" them in Higgsfield using models like Seedance 2.5, then manually compose everything in Premiere Pro/After Effects. I want as much creative control as possible. AI, in this case, only replaces "hands" of designers, animators. 3/ Human team ↳ The same script. I curate/direct my team to design, animate, edit, add SFX, manually. Figma + After Effects. Disclaimer: I understand that I have a clear advantage of being a creative director. My "AI results" might not be typical or achievable by an average person with 0 video experience (but let's see! haha) The goal is to see what modern tools are capable of! I will keep you updated almost daily. Challenge begins! follow along :)