Fully Funded PhD Studentship in Three-Channel Interview Assessment: Responsible Multimodal AI for Fair and Evidence-Based Hiring

Job title: Fully Funded PhD Studentship Package: Three-year studentship Job category/type: Research Fully Funded PhD Studentship in Responsible AI for Interview Assessment and Candidate-Organisation Compatibility (UK Home Applicants Only) Project: Three-Channel Interview Assessment: Responsible Multimodal AI for Fair and Evidence-Based Hiring Commercially Funded Doctoral Studentship in Collaboration with IMS Group , the University of Bedfordshire , and Luton AI . A full-time, fully funded PhD studentship for UK Home students is available at the University of Bedfordshire to develop and evaluate responsible artificial intelligence methods for interview assessment and candidate-organisation compatibility. The studentship is commercially sponsored by IMS Group and will connect doctoral research with a major international workforce-solutions business operating across recruitment, finance, data, marketing and managed IT services. The successful candidate will also become part of Luton AI, the University of Bedfordshire's applied AI ecosystem. Luton AI brings together academic research, specialist facilities, external partners and real-world projects to support the responsible development and practical application of artificial intelligence. Through the IMS Group-funded studentship, the successful candidate will receive full tuition-fee support, an annual stipend, academic supervision, access to research infrastructure and specialist facilities, doctoral training, and opportunities to engage with an international commercial partner. Funding The studentship will provide: Full Home University tuition fees for three years - please note, this opportunity is only available to UK Home students. An annual stipend for up to three years - confirmed amount to be inserted. Access to specialist research facilities, computing infrastructure and doctoral training. Academic supervision and support from the University's research community. The stipend will be awarded annually for up to three years, subject to satisfactory academic progression and in accordance with the funding agreement between IMS Group and the University of Bedfordshire. Key dates Closing date: Sunday 16 August 2026 Interview date: Virtual interviews will take place during the week commencing Monday 31 August 2026 Expected start date: October 2026 Study mode: Full-time Duration: Three years, subject to satisfactory progression The project Interviews remain one of the most widely used methods of personnel selection, yet decisions can be influenced by inconsistent judgement, unstructured questioning and the way interviewers interpret verbal and non-verbal behaviour. As recruitment becomes increasingly digital and AI-assisted, there is a pressing need for methods that are demonstrably valid, fair, transparent and acceptable to candidates. This PhD project will develop and empirically test a Three-Channel Interview Assessment Model combining: verifiable attributes such as qualifications, work samples and assessed skills; self-reported information such as experience, motivations and structured interview responses; and observable interaction signals such as gaze direction, facial movement, posture, gesture and vocal prosody. The successful candidate will investigate how human interviewers combine these channels, the implicit weight assigned to each source of evidence, and how effects such as halo, similarity bias and cultural interpretation may influence decisions. Controlled experiments and policy-capturing methods may be used to compare interviewer judgements with evidence-based outcome measures. The project will also explore machine-learning, multimodal data analysis, computer vision, audio analysis and explainable AI methods. Rather than assuming that behavioural signals reveal personality, deception or suitability, the research will test whether any signals provide reliable and incremental information, under what conditions, and with what limitations. Human, structured and AI-assisted assessment approaches will be compared using appropriate measures of predictive validity, reliability and calibration. Where feasible, the research may include longitudinal validation against outcomes such as performance, progression, retention and candidate experience. Fairness, privacy, informed consent, accessibility and human oversight will be central to the research design. Protected characteristics will not be used to determine candidate suitability; where demographic information is collected for approved research purposes, it will be used to identify and mitigate differential performance, bias or exclusion. The longer-term objective is to produce an evidence-based and auditable framework for responsible interview assessment that supports better workforce decisions without replacing professional judgement or reproducing historical inequalities. The research will consider routes to practical evaluation within recruitment and workforce environments relevant to IMS Group. Resea