If our work sounds interesting, we encourage you to apply. All applications are assessed on a rolling basis, so it is best to apply as soon as possible.
"AISI has built a wonderful group of technical, policy, and civil service experts focused on making AI and AGI go well, and I'm finding it extremely motivating to do technical work in this interdisciplinary context."
Geoffrey Irving
Chief Scientist
"AISI is situated at the centre of the action where true impact can be made. I'm excited about the opportunities unfolding in front of us at such a rapid pace."
Professor Yarin Gal
Research Director
Our typical interview process includes submitting a CV and short written statement, skills assessments such as a technical interview and a take-home coding test, and 2-4 interviews, including a conversation with a senior member of our team. We tailor this process as needed for each role and candidate.
Please note that if you're applying to our technical roles, this privacy policy applies.
This application is for those without a preference for a team. We prefer you apply to team-specific RE roles below. /// Design and build evaluations to assess the capabilities and safety of advanced AI systems. Candidates should have relevant experience in machine learning.
This application is for those without a preference for a team. We prefer that you applying team-specific RS roles below. /// Lead research projects to improve our ability to assess the capabilities and safety of advanced AI systems. Candidates should have relevant experience in machine learning.
Design experiments and build evaluations to assess the cyber offensive capabilities of advanced AI systems. Candidates should have relevant experience in machine learning and cybersecurity.
Drive projects to understand advanced AI systems' vulnerability to misuse. Candidates should bring experience in ML research, ML engineering, or in security (e.g. redteaming in other domains).
Lead research projects to improve our ability to assess the cyber offensive capabilities of advanced AI systems. Candidates should have relevant experience in machine learning and cybersecurity.
As a team manager, you'll be heading up a multi-disciplinary team including scientists, engineers and domain experts on the capabilities that we are evaluations. These include autonomous replication, AI R&D, manipulation and deception.
Build large-scale experiments, to empirically evaluate risks such as uncontrolled self-improvement, autonomous replication, manipulation and deception. Collaborate with others to push forward the state of the science on model evaluations.
Research risks such as uncontrolled self-improvement, autonomous replication, manipulation and deception. Improve the science of model evaluations with things like scaling laws for dangerous capabilities.
As an interpretability research scientist or engineer, you'll lead early work to push forward the science on detecting scheming and white-box evaluations.
Drive research to develop our understanding of how safety cases could be developed for advanced AI. You'll work closely with Geoffrey Irving to build out safety cases as a new pillar of AISI's work.
You will be a part of the Testing Team, which is responsible for our overall testing strategy, and the end-to-end preparation and delivery of individual testing exercises. You will collaborate closely with researchers and engineers from our evaluations workstreams, as well as policy and delivery teams. Your role will be broad and cross-cutting, involving project management, strategy, and scientific and policy communication.
As workstream lead of a novel team, you will build a team to evaluate and mitigate some of the pressing societal-level risks that Frontier AI systems may exacerbate, including radicalization, misinformation, fraud, and social engineering.
As workstream lead for this novel team, you will build and lead a multidisciplinary team to evaluate and mitigate the behavioural and psychological risks that emerge from AI systems. Your teams’ work will address how human interaction with advanced AI can impact human users, with a focus on identifying and preventing negative outcomes.
AISI is expanding our Systemic Safety team. This team is focussed on identifying and catalyzing interventions which could advance the field of AI safety and strengthen the systems and infrastructure in which AI systems are deployed. As the Workstream Lead for this team, you will build and lead a multidisciplinary team focussed on pushing systemic safety forward as an agenda and creating the global environment for responsible innovation.
As the Head of Security at the AI Safety Institute (AISI), you will lead on building a cyber resilient AISI. This will include efforts to harden our systems and protect our people, information and technologies. You think big picture about organisational risk based on mission objectives and a calibrated understanding of existing and potential attacks. You want to combine meaningful security with creative solutions rather than being limited to the compliance playbook.
AISI’s Science of Evaluations team will conduct applied and foundational research focused on two areas at the core of our mission: (i) measuring existing frontier AI system capabilities and (ii) predicting the capabilities of a system before running an evaluation.
As a Research Engineer in the LLM evaluations team of the Chem/Bio workstream, you will develop and run evaluations that measure the ability of LLMs to provide detailed end-to-end instructions and troubleshooting advice for biological/chemical tasks and/or automate key steps of the scientific R&D pipeline.
Successful candidates will work with other researchers to design and run studies that answer important questions about the effect AI will have on society. For example, can AI effectively change people's political and social views? Research Scientists/Engineers have scope to use a range of research methodologies and drive the strategy of the team.
The Testing Team is a high-profile and high -performing team in AISI, responsible for delivering against AISI’s core mission: developing and conducting evaluations on frontier AI systems, before and after they are deployed. - AISI is the world’s first state-backed organisation doing this work. We work closely with frontier labs, and our work on testing delivers impact to both companies and governments around the world by providing high quality empirical evidence of frontier AI capabilities and risks.
As a member of this team, you will use cutting-edge machine learning techniques to improve model performance in our domains of interest. The work is split into two sub teams: Agents and Fine-Tuning. Our Agents Team focuses on developing the LLM tools and scaffolding to create highly capable LLM-based agents, while our Fine-Tuning Team builds out finetuning pipelines to improve models on our domains of interest.
As a member of this team, you will use cutting-edge machine learning techniques to improve model performance in our domains of interest. The work is split into two sub teams: Agents and Fine-Tuning. Our Agents Team focuses on developing the LLM tools and scaffolding to create highly capable LLM-based agents, while our Fine-Tuning Team builds out finetuning pipelines to improve models on our domains of interest.
The AI Safety Institute (AISI) is looking for exceptionally motivated and talented Full-Stack Engineers to join our Platform Engineering team. Our platform is the core of our project to build and run safety evaluations for next-generation frontier AI systems: in this diverse role you’ll collaborate with our research teams and blend web development, UX engineering and data visualisation to provide inference channels, facilitate hosting our own models, and create expert interfaces for evals development.