Biologist, DeepMind

Overview

In this role you evaluate and mitigate safety risks of frontier AI models in biology, developing evaluation frameworks and red-teaming strategies. You collaborate with CBRNE and safety experts to block harmful outputs while preserving scientific utility. You will communicate complex risks and mitigation insights to leadership and external stakeholders, contributing to safety policy evolution. This position sits at the intersection of biology, AI safety, and policy, shaping responsible deployment of advanced models.

Responsibilities

• Design and execute biology-focused AI risk evaluations and red-teaming plans
• Develop frameworks to assess capability, safety, and uplift of AI models
• Collaborate with CBRNE engineers to balance safety with scientific utility
• Partner with AI ethics, policy, and safety experts to align risk mitigation with organizational standards
• Communicate evaluation results and mitigation strategies to leadership and external stakeholders
• Support development and refinement of safety policies and guidelines for frontier AI in biology
• Contribute to cross-disciplinary research and publications in the biology–AI safety nexus

Key requirements

• PhD in relevant biology field or equivalent practical experience
• 3 years of wet-lab experience with dual-use research, high-consequence pathogens, or biological toxins
• Experience utilizing LLMs or narrow-use AI models to analyze and mitigate biological risks
• Authored peer-reviewed publications in microbiology, virology, host-pathogen interactions, protein biochemistry, or immunology
• Collaborative mindset
• Clear and concise scientific communication
• Strong analytical thinking and problem-solving
• Experience with LLMs or AI evaluation for biology safety
• Knowledge of dual-use research issues and biosafety concepts
• Ability to identify misuse scenarios of emerging biotechnologies

Report this job