Threat Modeler Lead, CBRNE, DeepMind

Threat Modeler Lead, CBRNE, DeepMind Overview

At Google DeepMind, this role sits at the point where advanced AI capability meets serious responsibility. The Responsible Development and Innovation team works on measuring and evaluating AI risks in the chemical, biological, radiological, nuclear, and explosives domains, and this position is central to that effort. The person in this role will act as the operational lead for building, improving, and sustaining the threat models used to assess dual-use risks in Google DeepMind’s most advanced systems. Those models feed directly into the Frontier Safety Framework, helping shape risk calibration and informing decisions about whether a model is ready for release or deployment. This is a hands-on role for someone who can take in large amounts of technical and scientific information, identify what matters most, and move decisively in a fast-moving environment.

The work is framed as an opportunity to apply practical threat intelligence to frontier AI safety. The team describes itself as part of a pioneering AI lab with interdisciplinary expertise, focused on using AI to address major global challenges while keeping safety and ethics at the center of its work. The broader environment includes global collaboration, scientific discovery, and product innovation at scale.

What you will do

• Develop, update, and preserve the threat modeling framework, including the thresholds used to define critical risk levels, and connect that framework to Frontier Safety Framework review workflows that support deployment decisions.
• Work across functions to design and carry out evaluations that detect CBRNe-related risks in high-capability AI models.
• Partner with evaluation and mitigation teams to assess whether the safeguards proposed for advanced models are strong enough.
• Engage with outside stakeholders, including government bodies, external organizations such as the Frontier Model Forum, and subject matter experts outside Google.
• Keep track of the wider external landscape, including emerging dual-use techniques and developments across frontier AI.
• Operate as the person who can rapidly absorb complex scientific and technical information, identify blind spots, and execute with discipline under time pressure.
• Use real-world threat intelligence to support safety work for AI systems where the consequences of failure could be severe.

What the role requires

• A PhD in a scientific, engineering, data science, or related field, or equivalent practical experience.
• At least two years of experience in one of the following settings: a national laboratory, a government defense organization, a military intelligence unit, or a specialized research institution.
• Strong ability to translate highly technical and high-risk CBRNe issues into clear, actionable guidance for senior business leaders, governance bodies, and policy experts.
• Experience leading independent work and synthesizing large datasets in demanding, fast-paced environments with tight deadlines.
• Comfort using both qualitative and quantitative models to make decisions when information is incomplete or uncertain.
• Experience with red-teaming, evaluating, and using large language models to uncover systematic weaknesses and potential misuse scenarios.
• Knowledge of dual-use concerns in CBRNe domains, with particular emphasis on biological risks and laboratory or defense-related contexts.
• A demonstrated record of reviewing complex technical documents and quickly identifying the most important risks and vulnerabilities.
• English fluency, since Google says English proficiency is required for all roles unless the posting says otherwise.

Additional information from the posting

The role is based in London, UK. Google notes that it is an equal opportunity and affirmative action employer and says it is committed to building a workforce that reflects the users it serves and to maintaining a culture of belonging. The company also states that applicants who need an accommodation can use the accommodations form referenced in the posting.

Google further says that information collected through a Google Careers profile and any applications submitted is handled under the Applicant and Candidate Privacy Policy. It also includes standard hiring notices: Google does not accept agency resumes, does not want resumes forwarded to its jobs alias or employees, and is not responsible for fees tied to unsolicited resumes. The posting also explains that equity grants are handled solely and discretionarily by Alphabet Inc., are subject to approval and plan terms, and do not create any guaranteed entitlement.

To apply, use the Apply option on the Google Careers posting for this role.

Degree Requirement: Bachelor’s

Visa Sponsorship May be