Research Engineer, Frontier Safety Risk Assessment

Core responsibilities:

• Design, implement, and empirically validate approaches to assessing and managing catastrophic risk from current and future frontier AI systems
• Identify new risk pathways within current areas (loss of control, ML R&D, cyber, CBRN, harmful manipulation) or in new ones
• Conceive of, design, and develop new ways to measure pre-mitigation and post-mitigation risk
• Forecast and scenario plan for future risks which are not yet material

This role contributes to measuring and assessing potential catastrophic risks from current and future AI systems, ensuring safety and ethics remain the highest priority in AI development and helping prevent harmful outcomes from frontier AI technologies.

Required skills and experience:

• Extensive research experience with deep learning and/or foundation models (e.g., PhD in machine learning)
• Adept at generating ideas, designing experiments, and implementing these in Python with real AI systems
• Keen to address risks from foundation models and have thought about how to do so
• Strong, clear communication skills with ability to engage technical stakeholders

Benefits include: enhanced maternity, paternity, adoption, and shared parental leave; private medical and dental insurance for yourself and any dependents; flexible working options; healthy food; on-site gym; faith rooms; terraces; relocation support and immigration services; plus bonus, equity, and benefits in addition to salary.

Google DeepMind is a team of scientists, engineers, and machine learning experts working to advance artificial intelligence for widespread public benefit and scientific discovery, with safety and ethics as the highest priority.