PhD Scholarship in bias propagation in generative AI-driven decision systems and its implications for trust, governance, and regulatory compliance

PhD Scholarship in Bias Propagation in Agentic and Generative AI-Driven Decision Systems

Job No.:680086

Location:Caulfield campus

(with possible collaboration activities at Clayton campus)

Employment Type:Full-time

Duration:3-year fixed-term appointment(subject to satisfactory progress)

Remuneration:The successful applicant will receive a tax-free stipend at the current value of $37145 per annum 2026 full-time rate as per the Monash Research Training Program (RTP) Stipend for Trust Governance and Regulatory Compliance

This opportunity invites applications from outstanding domestic and international candidates who are interested in undertaking a PhD focused on bias propagation in agentic and generative AI-driven decision systems and its implications for trust governance and regulatory compliance. The project is part of an interdisciplinary research team led by experts from the Opportunity Tech Lab within the Monash Business School and the Faculty of IT at Monash University.

The PhD candidate will be working with a team of distinguished researchers:

• Professor Charmine Härtel (Department of Management Monash Business School; Director Opportunity Tech Lab)
• Professor Kristian Rotaru (Department of Accounting Monash Business School; Associate Director Opportunity Tech Lab; Chair of the Steering Committee Monash Business Behavioural Laboratory)
• Dr Mor Vered (Department of Data Science & AI Faculty of IT)
• Dr Estelle Wallingford (Department of Business Law & Taxation Monash Business School; Opportunity Tech Lab)

The Opportunity

This project addresses a pressing public policy and social issue: the propagation of bias in agentic and generative AI systems and its impact on human decision-making trust and regulatory design. The rapid evolution of AI from generative models that produce text and recommendations to agentic AI systems that autonomously plan act and make decisions with limited human oversight has transformed how critical choices are made across high-stakes domains including financial markets healthcare and legal practice. These systems capacity to produce biased yet seemingly neutral outputs poses a significant and growing risk to fairness accountability and public trust.

A core focus of the project is understanding how AI-generated explanations and autonomous AI actions influence trust formation cognitive effort and user behaviour including the risk of over-reliance (automation bias) or inappropriate scepticism when AI outputs are misleading hallucinated or generated through opaque multi-step reasoning. As agentic AI systems increasingly operate across organisational processes with minimal human intervention understanding how bias propagates through chains of autonomous decisions becomes essential.

The candidate will contribute to the development of empirically validated methods for identifying measuring and mitigating bias propagation effects. The research will involve experimental studies utilising the advanced neurophysiological infrastructure of the Monash Business Behavioural Laboratory (MBBL) including EEG eye-tracking pupillometry fNIRS and psychophysiological assessment alongside cognitive modelling and regulatory analysis. This combination of cutting-edge neuroscience methods with legal and governance scholarship is a distinctive feature of the project.

The project is situated within a dynamic and rapidly evolving regulatory Australia the Privacy Act reforms introducing new automated decision-making transparency obligations take effect in December 2026 the Australian AI Safety Institute becomes operational in early 2026 and ongoing policy development under the National AI Plan (2025) signals increasing regulatory attention to high-risk AI applications. Internationally the EU AI Act is moving into enforcement and jurisdictions worldwide are grappling with how existing legal frameworks apply to autonomous AI systems. The PhD candidate will have the opportunity to contribute to this critical policy discourse through empirically grounded research.

This doctoral project will conduct experimental research on generative and agentic AI systems with a focus on the experiences and risks faced by vulnerable populations. The research program is designed to produce empirically grounded insights that inform harm mitigation strategies transparency mechanisms and responsible deployment protocols relevant to both policymakers and organisational decision-makers. Rather than pursuing broad ethical theorising the project emphasises causal evidence on how AI design and deployment choices shape downstream social and economic outcomes for populations with asymmetric power information or risk exposure. We are looking for a researcher who wants to produce empirical evidence that informs real policy and organisational decisions rather than work that is mainly philosophical or purely technical.

Essential Skills and Experience

• A background in a relevant field such as behavioural science cognitive science data science psychology human-computer interaction law or a related discipline
• Demonstrated experience in empirical research (quantitative qualitative or mixed methods)
• Strong written communication skills
• A clear interest in interdisciplinary research on AI decision-making governance and ethics

Desirable skills

• Experience with experimental design and behavioural data collection
• Familiarity with generative or agentic AI systems explainability (XAI) or algorithmic fairness
• Skills in statistical analysis and/or coding (e.g. R Python C)
• Exposure to neurophysiological measurement methods (e.g. EEG eye-tracking pupillometry fNIRS)
• Interest or training in technology law digital regulation or AI ethics
• An ability to engage in legal research including familiarity with legislation regulations and case law would be advantageous but is not essential

To Apply

This position has a two-stage selection process.

Stage 1: Expression of Interest. To apply please submit an Expression of Interest (EOI) via email to Professor Kristian Rotaru (). Please use the following subject line: EOI PhD Scholarship AI Bias Your Full Name.

Your EOI must include the following:

• Curriculum Vitae (CV) including details of academic qualifications research experience publications (if any) and relevant professional experience
• Certified copies of academic transcripts from all tertiary qualifications
• Evidence of English language proficiency (where applicable e.g. IELTS TOEFL or equivalent)
• A brief research proposal (maximum 3 pages) that clearly aligns with the projects focus on bias in agentic and generative AI-driven decision systems. The proposal should outline your interest in the topic specify the theoretical and methodological approaches you are considering and describe how your academic background and prior experience equip you to contribute to the projects objectives. Where possible please highlight relevant skills such as experimental design data analysis neurophysiological methods or legal/regulatory analysis that can be directly cross-referenced against the projects interdisciplinary goals. The proposal should also reflect your enthusiasm for working at the intersection of business IT cognitive science and law.

Incomplete applications (e.g. EOIs submitted without a CV or academic transcripts) will not be considered.

Stage 2: Interview and Formal Application. Candidates shortlisted from the EOI stage will be invited to discuss their ideas via Zoom prior to submitting a formal PhD application to the Faculty of Business and Economics. The successful candidate will enrol in an interdisciplinary cross-faculty project with the PhD degree to be awarded by the Faculty of Business and Economics upon completion of the project and the Monash doctoral requirements.

Enquiries:Professor Kristian Rotaru

Applications Close: Friday 17 April 2026 11:55pm AEST

Supporting a diverse workforce