Keynote: Refik Anadol Studio: Rainforest AI Research
summary
Studio presents custom PyTorch models that generate Amazon fauna, flora and funga and a Large Nature Model trained on field captures from 16 sites plus drones and sonic and weather sensors, aiming to preserve Yawanawa language and culture. Work history includes 200+ GANs on 4 billion images, a 1.2 billion point Gaudí scan, 113 million social images for NYC, MoMA acquisition, and a $3.2M NFT sale from 1,000 generative pieces for the Yawanawa. Tools include PyTorch, diffusion models, and Firmenich’s scent AI.
AI-written summary · about
chapters (8)
- 0:00Refik Anadol Studio intro
- 1:25Data as memory & fluid dynamics
- 3:12AI meets architecture
- 5:56Machine hallucinations
- 7:47AI in public spaces
- 10:53Nature as advanced tech
- 12:57Large Nature Model
- 15:27Closing vision
related talks
39mKeynote: Artistic Applications of Artificial Intelligence
40mKeynote: Open-source Multimodal AI in the Wild
16mKeynote: Marin: An Open Lab for Frontier AI
33mThe Dawn Of Universal Creativity
32mKeynote: Powering Generative AI with Kubernetes
46mData Discussion - AI for Good
47mThe impact of Generative AI in different industries
12mKeynote: What's New in AI