Agentic AI Risk Modelling and Mitigations

Agentic AI Risk Modelling and Mitigations

London, UK

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.

The deadline for applying to this role is Sunday 8 March 2026, end of day, anywhere on Earth.

As AI systems grow more capable and autonomous, understanding how humans could lose the ability to oversee, correct, or shut down these systems becomes critical – as does identifying what we can do to prevent it. Risk models for AI agents (for example, loss of control risk models) remain far less developed than those in comparable domains like cybersecurity and chem‑bio, and practical mitigations remain under explored (especially beyond traditional alignment and control work).

AISI is building a new team to close this gap. The new Agentic AI Risk Modelling and Mitigations team will develop rigorous models of how agentic AI could cause harm, identifying practical mitigations with a focus on measures the UK government are well‑placed to implement. We will draw on expertise only available within government – especially the national security community – to develop risk models and mitigations far more developed than those in academia or industry.

The hiring manager for this role is Benjamin Hilton; the team is advised by Geoffrey Irving. You’ll collaborate closely with researchers across AISI's red teams, evaluation teams, and alignment team, as well as with government stakeholders.

Your work will draw on empirical evidence from AISI's evaluations, along with the broader cybersecurity and ML literature to develop detailed and precise threat models and mitigations. You’ll need to reason carefully about complex and uncertain scenarios and communicate findings clearly to both technical researchers and policy decision‑makers. Some projects may also involve hands‑on ML or cybersecurity work, in collaboration with government partners, to develop mitigations.

We are open to hires at junior, senior, staff, and principal research scientist levels. We may also make an offer to particularly promising candidates with management experience to lead the workstream in a management role.

• Developing detailed models of specific loss‑of‑control scenarios – such as deceptive alignment during internal deployment, or a long‑horizon agentic cyberattack – specifying their causal structure, key assumptions, and plausibility given current and projected AI capabilities and propensities.
• Translating risk models and associated uncertainties into specifications for AISI's red teams and evaluation teams – identifying the tests that would provide the most informative evidence about whether specific risk pathways are viable.
• Analyzing the effectiveness of mitigations – such as monitoring infrastructure, compute governance, deployment guidelines, or containment protocols – drawing on input from national security stakeholders, and assessing which risk pathways remain plausible once mitigations are in place.
• Collaborating and communicating with government and national security stakeholders to develop and implement possible interventions, in partnership.

In accordance with the Civil Service Commission rules, the following list contains all selection criteria for the interview process.

The experiences listed below should be interpreted as examples of the expertise we’re looking for, as opposed to a list of everything we expect to find in one applicant.

You may be a good fit if you have:

• Experience producing detailed threat models, risk analyses, safety cases, or similar structured analytical work – in AI…