About the Project
AI_CDT_DecisionMaking
Details
AI increasingly supports high-stakes decisions in areas such as healthcare, policing and social care. Effective human oversight requires people to challenge or override AI when appropriate, yet we lack methods to measure whether they genuinely exercise judgement rather than simply accept AI recommendations.
Creating AI systems that enhance rather than replace human capabilities requires effective oversight. Current approaches and benchmarks often overlook the unique context, experience and information people bring to real decisions. Better systems must support human agency and enable people to use this knowledge when assessing AI recommendations.
Building AI that better supports complex decision-making requires better evaluation: benchmarks in which the person holds information the system does not, so that overriding the AI is sometimes the right call; tasks that require complex evidence synthesis and have more than one defensible answer; and evaluation suites that score whether context, case history and professional experience were brought to bear, whether alternatives were examined, and whether otherwise identical cases were treated alike. Such evaluations test not only how capable a model is, but how capable it is at supporting human decision-making. As models become more powerful without becoming more helpful, their adoption is accompanied by alarming effects such as deskilling, loss of the expertise needed to oversee them, and a growing class of unpredictable and often undetectable errors. Without evaluation that rewards human agency, the default trajectory is AI that replaces our thinking rather than elevating how we think, create and work.
The student will join a group dedicated to building AI that genuinely enables human capability, by creating and evaluating AI systems that support high-stakes decision-making. We aim to move beyond passive verification and cognitive forcing functions towards genuine human-AI complementarity. This project will focus on developing ML evaluation suites and methodologies that measure whether an AI supports or undermines effective oversight, decision fairness and human agency, in ways that are valid, reproducible, and usable by developers, procurers and regulators. Depending on the student's interests, the project can take a more theoretical, technical, or applied direction.
The student will benefit from unique access to data from controlled and practitioner studies in the group and contribute to the emerging field of evaluation science for generative AI. There will be the option to engage with the group's partnerships with Police.AI, government and third-sector agencies.
Student Desirable Background:
A strong quantitative degree in computer science, machine learning, statistics, mathematics, engineering or a related field, with solid programming skills in Python. An interest in AI evaluation, human-computer interaction, cognitive science, or the societal and policy dimensions of AI is desirable; experience with evaluation work or human-participant studies is an advantage but not required. This project will work both for students from a more theoretical background who want to work on the formal side of evaluation, and students from an applied background who want to work closely with practitioners and real use cases.
By submitting an application for a University of Cambridge project, you consent to your documents being shared with the relevant supervisory team at the partner institution. If you have questions about this process [contact details available in the full listing] before applying.
How to apply:
Please apply under the University of Manchester application portal and select 'PhD in Artificial Intelligence' (OAA Application form)
When applying, you must include the project's full name, your supervisor's name, funding status, academic history, and two referees' contact details. Applications with missing documents will not be processed. Late or incomplete submissions will not be considered, and the university is not responsible for missed deadlines.
After you have applied you will be asked to upload the following supporting documents:
- Final Transcript and certificates of all awarded university level qualifications
- Interim Transcript of any university level qualifications in progress
- CV
- Supporting statement: A mandatory one- to two-page statement explaining your motivation for postgraduate research, why you chose Manchester, relevant academic or work experience, key research findings, and the techniques or skills you've gained. Applications will be put on hold if this is not provided.
- Contact details for two referees (Please ensure referee email addresses are official university or workplace accounts, as we may need to verify the references.)
- English Language Certificate (if applicable)
Equality, diversity and inclusion is fundamental to the success of The University of Manchester, and is at the heart of all of our activities. We know that diversity strengthens our research community, leading to enhanced research creativity, productivity and quality, and societal and economic impact. (Equality, diversity and inclusion | The University of Manchester)
We actively encourage applicants from diverse career paths and backgrounds and from all sections of the community, regardless of age, disability, ethnicity, gender, gender expression, sexual orientation and transgender status.
We also support applications from those returning from a career break or other roles. We are dedicated to supporting work-life balance and offer flexible working arrangements to accommodate individual needs. Our selection process is free from bias, and we are committed to ensuring fair and equal opportunities for all applicants.
We are dedicated to supporting work-life balance and offer flexible working arrangements to accommodate individual needs, including a part-time study option.
Join us and help grow the UK's national capability in artificial intelligence, part of the UK's modern industrial strategy. Everyone is welcome if you are passionate about uncertainty quantification, decision making with humans in the loop and decision making for machine learning systems and their applications in Science and Engineering, and looking to develop your research and innovation skills at the doctoral level to further your career in the AI sector.
Funding Notes
The UKRI AI CDT in Decision Making for Complex Systems is part of the TechFirst offering stipends of £30,000 for students eligible for home fee status. The start date is September 2027.
The Project will be based in University of Cambridge.
References
[1] M. Vaccaro, A. Almaatouq, and T. Malone. When combinations of humans and AI are useful: A systematic review and meta-analysis. Nature Human Behaviour, 8(12):2293–2303, 2024. https://doi.org/10.1038/s[contact details available in the full listing]
[2] G. Bansal, T. Wu, J. Zhou, R. Fok, B. Nushi, E. Kamar, M. T. Ribeiro, and D. Weld. Does the whole exceed its parts? The effect of AI explanations on complementary team performance. In Proceedings of the 2021 CHI Conference on Human Factors in Computing Systems, 2021. https://doi.org/10.1145/[contact details available in the full listing]
[3] Z. Buçinca, M. B. Malaya, and K. Z. Gajos. To trust or to think: Cognitive forcing functions can reduce overreliance on AI in AI-assisted decision-making. Proceedings of the ACM on Human-Computer Interaction, 5(CSCW1):1–21, 2021. https://doi.org/10.1145/3449287
[4] M. Lee, M. Srivastava, A. Hardy, J. Thickstun, E. Durmus, A. Paranjape, I. Gerard-Ursin, X. L. Li, F. Ladhak, F. Rong, R. E. Wang, M. Kwon, J. S. Park, H. Cao, T. Lee, R. Bommasani, M. Bernstein, and P. Liang. Evaluating human-language model interaction. Transactions on Machine Learning Research, 2023. https://arxiv.org/abs/[contact details available in the full listing]
[5] S. Bordt and U. von Luxburg. A bandit model for human-machine decision making with private information and opacity. In Proceedings of the 25th International Conference on Artificial Intelligence and Statistics (AISTATS), PMLR 151:7300–7319, 2022. https://proceedings.mlr.press/v151/bordt22a.html
[6] P. Z. Labedzka, D. Peters, J. J. Dudley, and M. Zilka. Human-AI interaction for time-critical sensemaking in missing persons investigations. In Proceedings of the 2026 CHI Conference on Human Factors in Computing Systems, 2026. https://doi.org/10.1145/[contact details available in the full listing]
[7] B. Butcher, M. Zilka, J. Hron, D. Cook, and A. Weller. Optimising human-machine collaboration for efficient high-precision information extraction from text documents. ACM Journal on Responsible Computing, 1(2):1–27, 2024. https://doi.org/10.1145/3652591
[8] L. Weidinger, I. D. Raji, H. Wallach, M. Mitchell, A. Wang, O. Salaudeen, R. Bommasani, D. Ganguli, S. Hooker, and W. Isaac. Toward an evaluation science for generative AI systems. arXiv preprint arXiv:[contact details available in the full listing], 2025. https://arxiv.org/abs/[contact details available in the full listing]
