Summary
Anyma’s ÆDEN concert visuals combine generative-AI concept exploration with a conventional human-led VFX pipeline, according to a technical account published by The Tulum Times. Matteo Milleri said Google Gemini was used to iterate early ideas, while artists later handled modeling, keyframe animation, dancer-based motion capture, simulation, lighting and rendering in tools including Houdini, Maya, Unreal Engine and Blender. The finished sequences appear to be pre-rendered video synchronized to music and stage cues rather than visuals generated live. VFX instructor Pablo Reche argued that low camera angles, slow movement, hard lighting and dark backgrounds create much of the apparent monumentality. He also said a widely shared Coachella shot extending beyond the stage screen was augmented for the video feed and was not visible that way to the in-person crowd. Milleri has separately denied that his final visuals are AI-generated.
Hana’s perspective
I find the distinction between generative ideation and final-frame production more useful than the usual “AI or not” argument. Here, Gemini appears to have accelerated visual exploration, while dancers, animators and VFX artists still carried the work through modeling, simulation, lighting and rendering. That does not make AI irrelevant; it places it upstream, where it can still shape the aesthetic before a human production pipeline makes the images. The larger lesson for animation teams is that spectacle often depends less on a single breakthrough tool than on disciplined staging: camera perspective, restrained motion, lighting and synchronization can sell scale more effectively than raw complexity. It is also a reminder to separate what an audience sees in a venue from augmented footage created for social media.
Are you a writer with a great story? We'd love to hear from you: APPLY HERE