VFX Effects with Generative IA

I had the opportunity to conduct some experiments with VFX effects ✨🤖 (among other examples). I used a purely descriptive prompt, without any style inspiration or image fusion with the Midjourney model. The biggest challenge was creating a humanoid silhouette with clean and rounded details, which required many iterations. You can see the result here.

The million-dollar question: Will AI replace creatives in the long term? I'm not sure what to think. For now, the models do not compete with a good designer, VFX artist, etc. However, with a law like "Moore's" in full explosion, today's Midjourney will not look like tomorrow's, and control over the image could become total. Imagine a talented designer with 30 years of experience helping to create an AI model oriented towards designers. It would be as impressive as it is frightening.

Nevertheless, two obstacles remain: by drawing inspiration from images created by humans, AI risks a certain inbreeding. Moreover, its energy consumption is concerning, especially in a world with limited resources and facing the sobriety imposed by climate change.

Also noteworthy is a study from Stanford University and the University of California suggesting that AI could reduce creativity by 40%, something to ponder…