Designing Effective Human-AI Collaboration for National Security Analysis

Eligibility: UK/International (including EU) graduates with the required entry requirements

Duration: Full-time – between three and three and a half years fixed term

Application deadline: 25 Aug 2026

Interview date: Will be confirmed to shortlisted candidates

Start date: January 2027

For further details contact: Professor Andrew Glazzard

Introduction

The project addresses a timely and under-studied question for government agencies: what are the fundamental needs and activities of professional analysts, and how can LLMs support those needs without distorting judgement, weakening accountability, or encouraging over-reliance on the tools? The challenge is not only to access information, but to construct meaning, test competing interpretations, and produce defensible judgements under conditions of uncertainty. Security studies research has long described this as a sensemaking problem involving information foraging, representation, hypothesis generation, and the conversion of insight into analytic products. This perspective matters because it changes how AI support should be researched. If national security analysis is fundamentally a process of sensemaking, then the relevant question is not simply whether an LLM can summarise or classify material. The deeper question is whether LLMs can support analysts in the activities that matter most: maintaining situational awareness, tracing relationships between items and actors, developing and updating hypotheses and recognising when confidence is or is not warranted.

Previous human-machine collaboration research suggests that analysts benefit when technology supports evidence handling, hypothesis formation, and structured reasoning, but that the central challenge remains one of sensemaking rather than automation. Newer research suggests that LLMs show promise for handling scale, retrieval, and synthesis, but remain limited in contextual judgement and robust reasoning across ambiguity. The proposed studentship therefore shifts the focus of investigation away from the tool and towards the analyst. It will study natinal security analysis as a socio-cognitive practice involving information foraging, representation, hypothesis generation, and the conversion of insight into analytic products in environments characterised by ambiguity, data overload, fragmented evidence, and shifting time pressures. It will identify the points at which analysts need support, and evaluate whether LLMs can meet those needs in ways that improve analytical performance and organisational resilience. The result will be a theoretically grounded and operationally relevant account of where LLMs fit into national security work, where they do not, and how support should be designed around the analyst rather than around the model.

Current debate often moves too quickly to the capabilities of AI systems, especially interface features or explanation mechanisms. Those matter, but they are secondary to a more foundational issue: understanding the analyst’s cognitive work, institutional responsibilities, and practical constraints. This studentship is designed to address that fundamental gap.

This PhD is a cotutelle PhD between Coventry University (UK) and Stellenbosch University (South Africa). The successful candidate will primarily be based in Coventry, but will be required to undertake a significant period of study in South Africa as a requirement of the PhD.

Project details

The studentship will aim to identify the core needs, activities, and cognitive demands of security analysts, and to determine how LLMs can support those needs in ways that strengthen rather than displace human sensemaking and judgement.

Its objectives will be to:

Produce a rich account of security analysis as work, with particular attention to sensemaking, uncertainty, and judgement
Develop a task and needs taxonomy of security analysis grounded in empirical study of analysts and analyst-like expert populations
Examine where LLM capabilities align with actual analytical needs, such as retrieval, summarisation, continuity support (in shift-work settings) and structured exploration of evidence
Identify areas where LLM support is likely to be unhelpful or risky, especially where contextual interpretation, responsibility, and evidential defensibility are central
Deliver a human-centred framework for evaluating and governing LLM support in intelligence settings.

The proposed research questions (RQs) are therefore;

RQ1: What are the principal activities, constraints, and cognitive burdens involved in national security analysis as a form of professional sensemaking?
RQ2: Which analyst needs are most amenable to LLM support, and which remain fundamentally dependent on human expertise, contextual judgement, and institutional responsibility?
RQ3: How do LLMs affect key analytical practices such as information foraging, hypothesis generation, evidential updating, uncertainty management, and analytic communication?
RQ4: What risks arise when LLM support is misaligned with analyst needs, including automation bias, de-contextualisation, loss of provenance and weakened accountability?

Academically, the studentship will contribute to at least three developing fields – security studies, human-computer interaction and AI – by grounding discussion of LLMs in the realities of analytical work rather than in abstract claims of capability. Operationally, it will help government organisations make more disciplined decisions about where LLMs add value, where they create risk, and how to preserve human judgement in high-consequence environments. For policy makers, it can inform guidance on trustworthy adoption of generative AI in public-sector analysis, especially where evidential defensibility, auditability, and institutional accountability are essential.

Funding

Tuition fees and bursary.

Benefits

The successful candidate will receive comprehensive research training including technical, personal and professional skills. All researchers at Coventry University (from PhD to Professor) are part of the Doctoral and Researcher College, which provides support with high-quality training and career development activities.

Entry requirements

A minimum of a 2:1 first degree in a relevant discipline/subject area with a minimum 60% mark in the project element or equivalent with a minimum 60% overall module average.

PLUS

A master's degree in a relevant subject area.The master's must have been attained with minimum overall marks at merit level (60%)*. In addition, the dissertation or equivalent element in the master's must also have been attained with a minimum mark of merit level (60%).
The potential to engage in innovative research and to complete the PhD within 3.5 years.
A minimum of English language proficiency (IELTS academic overall minimum score of 7.0 with a minimum of 6.5 in each component).

Essential skills and knowledge

A first-class or upper second-class honours degree (or equivalent) in Computer Science, Artificial Intelligence, Human-Computer Interaction, Psychology, Cognitive Science, Information Science, Security Studies, Data Science, Social Science, or another relevant discipline.
An interest in the interaction between humans and artificial intelligence, particularly the use of generative AI in complex decision-making.
Strong analytical and critical thinking skills, with the ability to synthesise complex information and develop well-reasoned arguments.
An understanding of qualitative and/or quantitative research methods and a willingness to apply empirical methods to real-world research questions.
Excellent written and verbal communication skills, with the ability to present complex ideas clearly to both technical and non-technical audiences.
The ability to work independently while contributing effectively within a multidisciplinary research environment.
Strong organisational skills and the ability to manage a long-term research project.

Desirable knowledge and experience

Familiarity with large language models, generative AI, or natural language processing.
Knowledge of human-computer interaction (HCI), human-centred AI, cognitive systems engineering, or user-centred design.
Understanding of intelligence analysis, security analysis, decision-making under uncertainty, or sensemaking.
Experience conducting interviews, surveys, observations, workshops, or other empirical research involving human participants.
Experience analysing qualitative or mixed-methods data.
Familiarity with experimental design and evaluation methods.
Experience using generative AI tools to support research or analytical tasks.
Knowledge of AI ethics, responsible AI, explainability, or trust in AI systems.
Awareness of organisational change, technology adoption, or socio-technical systems.

Personal attributes

Curious about how people work with AI rather than simply how AI systems work.
Motivated to undertake interdisciplinary research spanning AI, human behaviour, and national security.
Able to think critically about emerging technologies and evaluate evidence objectively.
Comfortable engaging with practitioners and translating research findings into practical recommendations.
Strong interpersonal skills and the confidence to conduct interviews and facilitate discussions with stakeholders.
Open to learning methods from multiple disciplines, including computing, behavioural science, and organisational research.
Committed to conducting rigorous, ethical, and reproducible research.
Able to handle complex, ambiguous problems where there may be no single "correct" answer.

How to apply

To find out more about the project, please contact Professor Andrew Glazzard.

All applications require full supporting documentation, a covering letter, plus a 2000-word supporting statement showing how the applicant’s expertise and interests are relevant to the project.

Apply to Coventry University