Research
What makesan AI systemtrustworthyafter deployment?
Four territories, one question: what has to be true, technically and institutionally, before an AI system can be relied on where it affects people.
01 / Territory
AI assurance and accountability
What evidence about an AI system is adequate for a particular decision, held by a particular role, in a particular deployment?
Methods
- Reproducible public-records analysis
- Executable obligation reasoning
- Evidence adequacy scoring
- Adversarial stress testing
Selected work
02 / Territory
Privacy-preserving collaborative AI
What has to hold, technically and institutionally, for federated analysis across sensitive datasets to be trustworthy rather than merely decentralised?
Methods
- Federated learning
- Differential privacy
- Attribution auditing
- Governance design
Selected work
03 / Territory
Human-centred adaptive systems
How can adaptive and affect-aware systems support learning and wellbeing without overreading sparse, intimate signals?
Methods
- Human-computer interaction
- Time-series modelling
- Affective computing
- Design-based research
Selected work
04 / Territory
Data governance for public-interest systems
How should public institutions govern data about people who cannot meaningfully consent, and what does that require of the systems built on it?
Methods
- Policy analysis
- Institutional capacity assessment
- Comparative governance review
Selected work
Research projects
2025-
An assurance toolchain for deployed AI
A connected set of reproducible tools for deployment-conditioned risk analysis, executable regulatory reasoning, evidence adequacy scoring and adversarial stress testing.
2021-2024
Wearable data and clinician wellbeing
Doctoral research on whether wearable sensor data can support mental-health risk monitoring for frontline healthcare professionals, and where the method breaks down.