Research Scientist, Propensity, Cyber and Autonomous Systems Team

The AI Security Institute is the world's largest and best-funded team dedicated to understanding advanced AI risks and translating that knowledge into action. We’re in the heart of the UK government with direct lines to No. 10 (the Prime Minister's office), and we work with frontier developers and governments globally.

We’re here because governments are critical for advanced AI going well, and UK AISI is uniquely positioned to mobilise them. With our resources, unique agility and international influence, this is the best place to shape both AI development and government action.

Role Summary

Within the Cyber & Autonomous Systems Team (CAST) at AISI, the Propensity project studies unprompted or unintended model behaviour, particularly potentially dangerous behaviour: the propensity of a model to cause harm. Our current project is to study effect sizes of environmental factors on these propensities, e.g. whether models are consistently more willing to take harmful actions when their existence is threatened.

We build on previous work on this field by scaling to a range of different scenarios and variations, and looking particularly for effects that are consistent throughout.

Understanding model propensities is the key missing pillar in our overall picture of risk from autonomous AI. We already know that models have sufficient knowledge and ability to assist criminal users in conducting cyberattacks and causing significant harm. If they can also spontaneously develop the inclination to cause harm unprompted, the nature and scale of the threat is transformed. To justify a response sufficient to address this unprecedented threat, we need empirical evidence with strong scientific credibility.

In CAST within AISI, through our relationships with the rest of the UK government and national security apparatus (and their relationships with international counterparts), we have a unique ability to understand what they need and get it in front of them.

What we are looking for

The Propensity project team currently consists of one research scientist and two research engineers. We’re looking to add a second research scientist to help with challenges like those above, through discussion, written plans and designs, and writing or reviewing code that implements those designs. We expect that the strength of our answers to questions like these are likely to be a key factor in the strength of the conclusions we can draw, the claims we can back, and the accuracy of our predictions on which we rest the credibility of the work.

You would add the capacity we need to give our answers the next layer of depth and sophistication.

The ideal candidate will have the following skills:

• A proven ability to identify and ope rationalise key uncertainties in a research area, and propose and improve on experimental approaches for collecting evidence on these uncertainties,
• Knowledge of and experience in selecting and applying statistical inference methods in order to draw risk-relevant and action-guiding conclusions from experimental evidence,
• Ability to engage critically with existing or proposed research methodology, assessing to what extent such critiques impact the central conclusions of the work, and how a proposal could be adapted to address them,
• Strong enough Python knowledge to get hands-on with developing and iterating on our Inspect tasks (though Inspect itself can be learned on the job),
• A sufficient understanding of transformer architecture and training dynamics to inform interpretations and predictions of their observable behaviour (how output is sampled, the loss function used for pre-training, the differences between pre-training and post-training, what inference-time compute scaling is, etc.) – hands-on experience with MLE tasks like fine-tuning or RL is not required.

We expect these skills will be held by people with:

• 3+ years of experience in a quantitative research discipline (e.g. as a PhD student or data scientist or researcher) involving experimental design and analysis,
• Experience writing Python code meeting quality standards, e.g. in production environments or in collaboration with others,
• Professional or educational (or…