AI safety & policy
Full-time
Routes to impact
Direct high impact on an important cause
Description
Conduct technical research and build safety measures for advanced AI systems, focusing on risks from recursive self-improvement and loss of control.
- Develop scalable oversight, automated auditing, and model behavior evaluations
- Stress-test monitoring systems for scheming and other misalignment risks
- Build prototypes and integrate successful approaches into safety pipelines
- Coordinate technical safety work across teams and external communications
This text was generated by AI. If you notice any inconsistencies, please let us know using this form.
