Nuclear AI Safety Research Scientist

Google LLC; Mountain View (CA); On-site; USD 174,000 - 252,000; Full time

Job summary

Google DeepMind seeks a Nuclear Subject Matter Expert to evaluate and mitigate safety risks of frontier AI models in the nuclear domain, focusing on fission, fusion and radiative-hydrodynamics topics in a non-export controlled, unclassified framework.

The role acts as a technical authority within the ReDI team, coordinating with engineers to design safe model capabilities, develop evaluation datasets, and craft mitigations while communicating findings to leadership and external stakeholders.

Qualifications

• PhD in plasma physics, nuclear engineering, a related field, or equivalent practical experience.
• Experience collaborating with US national labs or agencies like IAEA, NNSA, or DARPA, or similar government/academic consortia.
• Experience in research/policy in stockpile stewardship, nonproliferation, arms control, or threat reduction.
• Technical experience in high energy density physics, inertial confinement fusion, implosion dynamics, criticality, missiles, or reactor operations, nuclear infrastructure.

Responsibilities

• Construct threat models prioritizing vulnerabilities, threat actors, and scenarios related to fission, fusion, thermonuclear weapons development and high energy density dual use regimes.
• Create robust evaluation datasets of complex physics and nuclear engineering prompts to elicit the maximum capability of the model to determine how to uplift malicious actors across the domain.
• Collaborate with engineers to design mitigations that block harmful outputs without overly degrading the model's scientific utility.
• Support development, refinement of Google’s nuclear safety policy. Communicate complex risks, evaluation findings, mitigation strategies to leadership and technical/non-technical external stakeholders.

Skills

Nuclear physics

Policy research

AI safety

Threat modeling

Mathematical modeling

Python

C++

MHD

Education

PhD in plasma physics or nuclear engineering

Equivalent practical experience

Tools

Python

C++

Radiation hydrodynamics

MHD

Job description

Google DeepMind seeks a Nuclear Subject Matter Expert to evaluate and mitigate safety risks of frontier AI models in the nuclear domain, focusing on fission, fusion and radiative-hydrodynamics topics in a non-export controlled, unclassified framework.

The role acts as a technical authority within the ReDI team, coordinating with engineers to design safe model capabilities, develop evaluation datasets, and craft mitigations while communicating findings to leadership and external stakeholders.