Research Scientist, FSF Risk Modeling and Governance, DeepMind

Research Scientist, FSF Risk Modeling and Governance, DeepMind Overview

We are seeking a Research Scientist to spearhead governance research and risk modeling efforts for the Frontier Safety Framework (FSF) at DeepMind. This role is crucial for addressing potential severe risks associated with frontier AI models, ensuring that our safety frameworks evolve to meet new challenges.

Responsibilities

• Lead governance research and risk modeling initiatives for the Frontier Safety Framework (FSF).
• Develop and update the FSF to address evolving safety, governance, and policy challenges.
• Conduct research and model pre-mitigation risks to inform empirical risk assessments and safeguard development.
• Collaborate with evaluation and mitigation teams to evaluate the effectiveness of safeguards in reducing risks.
• Manage a team of researchers and cross-functional collaborators on various projects.
• Represent the FSF risk assessment program in both internal and external governance discussions.
• Draft external safety reports and work with policy and governance teams to enhance the frontier AI governance ecosystem.

Requirements

• PhD in Economics, Philosophy, or a STEM discipline, or equivalent practical experience.
• At least 3 years of research experience in areas related to frontier AI safety or governance, such as threat modeling or safety frameworks.
• Proven ability to communicate complex technical concepts to diverse audiences, including governance bodies and policy experts.
• Demonstrated history of managing complex projects, particularly in research environments.
• Experience applying conceptual and quantitative reasoning to uncertain or under-constrained research problems.

Benefits

• Competitive salary ranging from $207,000 to $300,000 (USD) plus a target bonus of 20%.
• Equity options available.
• Comprehensive benefits package.

This position is not eligible for visa sponsorship.

Degree Requirement: Bachelor’s

Visa Sponsorship May be