Longterm Wiki

Redwood Research

Safety Org

Also known as: Redwood

Founded Jun 2021 (4 years old) by Nate Thomas, and Buck Shlegeris

A nonprofit AI safety and security research organization founded in 2021, known for pioneering AI Control research, developing causal scrubbing interpretability methods, and conducting landmark alignment faking studies with Anthropic.

Key People
2
2 total tracked
Funding Received
$27M
4 grants

Other Data

Major research initiatives and focus areas.
2 entries
NameDescriptionTeam SizeStarted
AI ControlDeveloping techniques to safely deploy AI systems even if they are not fully alignedJan 2023
Adversarial RobustnessTesting and improving robustness of AI safety techniques against adversarial inputsJun 2021

Key People

2
CEO & Co-founder
Jun 2021 – present
Previously researcher at MIRI; leads Redwood's research agenda on AI control
Nate ThomasFounder
Co-founder
Jun 2021 – present

Related Organizations

1
Aug 2022

Facts4

Organization
Founded DateJun 2021
HeadquartersSan Francisco, CA
Legal StructureNonprofit research lab
People
Founded Bynate-thomas, Buck Shlegeris
View all facts in KB explorer →

Funding Received

4
Total tracked: $27M
GrantFunderAmountDate
Redwood Research — General Support (2022)sourceCoefficient Giving$10.7 million2022-08
Redwood Research — General Support (2021)sourceCoefficient Giving$9.4 million2021-11
Redwood Research — General Support (2023)sourceCoefficient Giving$5.3 million2023-06
Redwood Research — AI Safety Research CollaborationssourceCoefficient Giving$1.1 million2025-05