Cristóbal Valenzuela Barrera

I'm a developer living in Brooklyn, NY. I'm the co-founder and CEO of Runway.

I was a researcher at ITP NYU Tisch School of the Arts. I've worked on open-source projects, contributed to AI research, and also made some art. I was born and raised in Chile. I'm primarily interested in building tools to augment human creativity. If that's interesting, join us. We are hiring.

Posts and Essays

• New Media
• Creative Software 2.0
• World Building
• Creative Meritocracy
• The Director's Cut
• A gym for the mind
• We're solving graphics backwards
• Expectations vs Reality
• Something else
• Taming the latent space
• We need more artists doing research
• The artist's revenge
• Sameness
• Art is about asking questions
• Oversimplify
• The medium is not the goal

Interviews and Talks

• AI Tools for the Creative Industry Bloomberg Screentime, October 9th, 2024.
• Instant Videos Could Represent the Next Leap in A.I. Technology The New York Times, April 4, 2023.
• ‘Not Everyone Is Trying to Build God’ The Information, August 4, 2023.
• Workshop on Machine Learning for Creativity and Design NeurIPS 2023, Dec 10, 2023.
• The Tools of Generative Art, from Flash to Neural Networks Art in America, January 8, 2020.
• AI could usher in new ‘golden era’ of cinema Semafor, Aug 30, 2023

c [at] runwayml.com - Twitter - Linkedin - Github

Previous projects

• 2018 / ml5.js ml5.js aims to make machine learning approachable for a broad audience of artists, creative coders, and students.
• 2017 / The Alternative Late Show. Experimenting with Image-to-Image translation models (pix2pix) for character animations.
• 2017 / Text 2 Image. Interactive real-time Text-to-Image generation using Attentional Generative Adversarial Networks (AttnGAN).
• 2017 / Uncanny Road. A tool for collaboratively hallucinating a never-ending road using Generative Adversarial Neural Networks.
• 2017 / Scenescoop. A tool to find semantically similar scenes between any two films or videos.
• 2017 / Selected Stories. A generative text editor that uses four custom LSTM networks running in the browser to generate text.
• 2016 / Sequential Stories. An iOS app that can write fictional stories using only a Text-to-Image model that captions images.
• 2016 / Word2Beat. A tool that maps word embeddings or a word space vector to sound beats.
• 2012 / Regression. An interactive installation that uses a typewriter to create text based generative videos, T2V.