Welcome to the Latent Dystopia
summary
Shows working Stable Diffusion runs that are patched to dump refinement steps, generates avatars and surreal scenes like Megatron training at a gym, exposes a Photoshop plugin workflow for masked edits, and documents batch renders on an EC2 $3/hr instance producing eight images per minute and thousands of images over a day. Demonstrates unstable outputs (extra fingers, merged faces), sketches risks: realistic deepfakes, copyright provenance from training sets, and GitHub Copilot–style code suggestions that reproduce security bugs.
AI-written summary · about
chapters (4)
related talks
29mIntegrating Stable Diffusion Generative AI in Artist Workflows
47mAI Image Generation: Visual Artistry with Midjourney, Adobe Firefly, and DALL-E3
36mArt in the Age of Algorithms
01h 03mIt's All About AI
52mThe Yoga of Image Generation with Stable DIffusion and ComfyUI
45mHey, Look, I'm an Artist Now © ML Image Generator
36mDesigning the Future: Stable Diffusion & the Rise of AI-Driven SaaS
45mGenerative AI for Image and Video Production