Redwood Research
The Redwood Research stream is looking for fast empirical iterators and strategists to work on control research.
Stream overview
- High-stakes control
- Diffuse control
- Model organisms
- Generalization and alignment
- AI futurism
Mentors
Alek Westover is a Member of Technical Staff at Redwood Research. He studied CS and Math at MIT, and previously did research in theoretical computer science.
Alex Mallen is a Member of Technical Staff at Redwood Research. He studied CS at the University of Washington, and previously worked at EleutherAI.
Aryan Bhatt is a Member of Technical Staff at Redwood Research. He studied Math and CS at Hunter College, and attended MATS in 2023. He studies AI Control.
Buck is the CEO of Redwood Research.
Cody Rushing is a Member of Technical Staff at Redwood Research. He studied CS at UT Austin before attending MATS in 2023.
James Lucassen is a Member of Technical Staff at Redwood Research. He studied CS at Harvey Mudd College, and previously did AI safety research at MIRI and CMU.
Julian Stastny is a Member of Technical Staff at Redwood Research. He has a Master's in ML from the University of Tübingen, and was previously a researcher at the Center on Long-Term Risk.
Ryan Greenblatt is Chief Scientist at Redwood Research. He studied Math and CS at Brown, and researches AI deception and control.
Tyler Tracy is a Member of Technical Staff at Redwood Research. He studied CS at the University of Arkansas, for Bachelor's and Master's degrees. He previously worked as a software engineer before attending MATS as a scholar himself.
Vivek Hebbar is a Member of Technical Staff at Redwood Research. He attended Stanford before attending MATS in 2022, and researches AI control.
Mentorship style
Depending on the mentor:
- 30-60 min weekly meeting
- potentially daily stand-ups
- Slack messages
Representative papers
the Ctrl-Z paper, the original control paper, the alignment-faking paper
Scholars we are looking for
We are looking for people who are:
- fast at empirical ML iteration.
- thoughtful and articulate about AI safety.
- strong at quantitative reasoning.
Redwood scholars, Redwood employees
Project selection
We will assign projects by default but are open to getting pitched on projects.
Community at MATS
MATS Research phase provides scholars with a community of peers.
.webp)
During the Research phase, scholars work out of a shared office, have shared housing, and are supported by a full-time Community Manager.
Working in a community of independent researchers gives scholars easy access to future collaborators, a deeper understanding of other alignment agendas, and a social network in the alignment community.
Previous MATS cohorts included regular lightning talks, scholar-led study groups on mechanistic interpretability and linear algebra, and hackathons. Other impromptu office events included group-jailbreaking Bing chat and exchanging hundreds of anonymous compliment notes. Scholars organized social activities outside of work, including road trips to Yosemite, visits to San Francisco, and joining ACX meetups.