Executive Summary
Generated July 30, 2026
Artificial Intelligence
Redwood Research
Strategic importance is our editorial rating of how central this entity is to the technology landscape we track; confidence reflects how well-sourced and current the underlying evidence is.
2
Chokepoint Score
2 other entities structurally depend on this one
Cooling
Activity
0 recorded changes in the last 90 days
Headquarters
Berkeley, California, USA
Website
Employees
30 (2026, approximate)
Founded
2021
Listing
nonprofit
Overview
A technical AI safety research organization focused on 'AI control' -- techniques to ensure AI systems cannot subvert safety measures even if misaligned -- advising Google DeepMind and Anthropic on assessing and mitigating risks from misaligned AI agents. Redwood Research partnered with the UK AI Security Institute to publish 'A sketch of an AI control safety case,' describing how developers can construct a structured argument that models are incapable of subverting control measures, and hosted ControlConf 2026, concluding that control techniques are now load-bearing for the safety of real-world agent deployments.
Strategic Connections
- UK AI Security Institute (AISI): Redwood Research partnered with the UK AI Security Institute to publish 'A sketch of an AI control safety case' for assessing whether models can subvert control measures.
- AI Safety, Alignment & Governance: Redwood Research partners with the UK AI Security Institute on AI control safety cases and advises Google DeepMind and Anthropic on misalignment mitigation.
- METR: Redwood Research and METR both conduct independent technical AI safety evaluation and control research, often collaborating with the same frontier labs and government institutes.
Strategic Risks
- AI control techniques remain nascent relative to the pace of frontier model capability growth, creating a persistent gap between deployed safeguards and model capability
- Reliance on continued voluntary cooperation from frontier labs (Google DeepMind, Anthropic) for research access could be curtailed at any time
Future Outlook
- Continued development and real-world adoption of AI control techniques as frontier labs build out initial control measures
- Expanding government partnerships (following the UK AISI collaboration) as more countries stand up AI safety evaluation capacity
Sources
1 source, 48% average source confidence — full citations in the Full Analyst Report.
Generated by InsightNodes — Technology Intelligence Platform. This report reflects sourced, evidence-backed information as of the dates cited above and is not investment advice.
Viewing the Executive Summary — switch to the Full Analyst Report.