Expression of Interest - Chem Bio
UK government research organisation that evaluates advanced AI risks and develops and tests mitigations to inform governments.
Funding history
Investors
About The AI Security Institute (AISI)
The AI Security Institute is a research organisation within the UK Department for Science, Innovation and Technology. It tests leading AI systems, conducts technical research on advanced-AI risks and solutions, develops evaluation infrastructure, and informs UK and allied-government policy.
Skills
Candidate Availability
Required and preferred rules are kept separate and reflect the wording in the original posting.
About the Role
You may contribute to research on chemical and biological risks associated with frontier AI systems. You could design and run model evaluations, develop benchmarks and experimental protocols, build tools for elicitation and monitoring, investigate safeguards, conduct threat modelling, and translate findings into recommendations for AI companies and government stakeholders.
Requirements
- Strong technical, scientific or research background relevant to chemical and biological AI risks
- Experience in machine learning evaluation, robustness, interpretability or capability assessment
- Experience in computational biology, bioinformatics, molecular biology, synthetic biology or related life sciences
- Experience in chemistry, computational chemistry, cheminformatics or related disciplines
- Experience in biosecurity, chemical security, threat assessment or dual-use risk analysis
- Experience in research engineering, scientific software, automation or technical research infrastructure
- Experience in experimental design, benchmarking or empirical evaluation
- Pass a criminal record check and obtain BPSS clearance before appointment
- Eligibility for CTC clearance is strongly preferred
Responsibilities
- Design and run evaluations of chemical and biological capabilities in frontier AI systems
- Develop benchmarks, datasets and experimental protocols for assessing high-consequence risks
- Study how model capabilities generalise or transfer across scientific domains and task settings
- Build tools and automated methods for capability elicitation, testing and monitoring
- Investigate model safeguards and identify weaknesses requiring strengthening
- Conduct threat modelling and risk analysis of realistic pathways to harm
- Translate technical findings into actionable recommendations for AI companies and government stakeholders
- Work with external science, security and policy partners to improve understanding of chem/bio-related AI risks
Benefits
- Pre-release access to multiple frontier models and ample compute
- Extensive operational support
- Five days off and annual learning and development stipends
- Funding for conferences and external collaborations
- Modern central London office and, where applicable, government offices in Birmingham, Cardiff, Darlington, Edinburgh, Salford or Bristol
- Hybrid working and flexibility for occasional remote work abroad
- Work-from-home equipment stipends
- At least 25 days of annual leave, 8 public holidays, extra team-wide breaks and 3 volunteering days
- Generous paid parental leave
- Employer pension contribution of 28.97% of base salary
- Cycling, donation, retail and gym discounts and benefits