02 / Research topics
Potential PhD Research Topics in Digital Trust
Explore suggested topics and the methods, tools and expertise that could support each investigation.
1
Auditable Digital Trust Frameworks for Generative Artificial Intelligence Systems
Despite rapid adoption of generative AI, organizations lack reliable ways to verify provenance, safety, privacy, accountability, and compliance across complex model supply chains. The gap is fragmented auditing and weak evidence standards. This research could create an auditable trust framework combining technical measurements, organizational controls, and stakeholder perceptions. Objectives include defining assurance indicators, developing continuous audit mechanisms, validating them across sectors, and examining how audit results influence adoption and regulatory decisions.
Applications and use cases
2
Privacy-Preserving Digital Identity Architectures for Inclusive Public Services
As digital identity becomes foundational to public services, centralized architectures expose citizens to surveillance, exclusion, identity theft, and function creep. Existing decentralized identity proposals often underaddress usability, recovery, governance, and unequal access. The study should design and evaluate privacy-preserving identity architectures using selective disclosure and verifiable credentials. Objectives include measuring trust across citizens and institutions, testing resilience and accessibility, and producing governance recommendations for interoperable national systems.
Applications and use cases
3
Measuring Public Trust in Algorithmic Decision Making Across Cultural Contexts
Current research frequently treats trust as a universal attitude, although perceptions of algorithmic decisions vary with culture, institutional history, power, and personal experience. The unresolved gap is the absence of validated cross-cultural measurement models that distinguish justified trust from blind acceptance. This project should develop a multidimensional scale, test measurement equivalence, and explain how transparency, contestability, human oversight, and fairness affect trust. Findings can guide culturally responsive AI governance and interface design.
Applications and use cases
4
Quantum-Resilient Cryptography for Long-Term Digital Trust Infrastructure
Emerging quantum capabilities threaten public-key cryptography supporting identities, certificates, payments, and archived records, while migration remains technically and organizationally immature. The key gap concerns trustworthy transition strategies that preserve interoperability, performance, and evidence integrity during hybrid deployment. This research should compare post-quantum schemes, develop migration risk models, and evaluate crypto-agility architectures. Objectives include quantifying sector-specific exposure, designing transition controls, and demonstrating verifiable trust continuity in realistic infrastructure environments.
Applications and use cases
5
Decentralized Data Provenance Systems for Trusted Artificial Intelligence Supply Chains
This project addresses uncertainty about where training data, models, updates, and outputs originate, especially when supply chains cross organizations and jurisdictions. Existing provenance solutions can be incomplete, expensive, or vulnerable to false claims by trusted insiders. The novelty lies in combining tamper-evident provenance with independent validation, privacy protection, and usable evidence for downstream decisions. Objectives are to model provenance threats, develop interoperable mechanisms, evaluate overhead, and establish how provenance evidence supports accountability and procurement.
Applications and use cases
6
Trustworthy Autonomous Agents Through Human Oversight and Verifiable Action Policies
Beyond conversational systems, autonomous agents can plan, negotiate, purchase, and operate software with limited supervision, creating novel risks of manipulation, goal drift, and unauthorized action. Current work emphasizes capability more than dependable delegation and recoverability. This research should develop verifiable action policies, adaptive oversight, and calibrated user controls. Objectives include formalizing agent trust boundaries, testing intervention strategies, measuring user understanding, and demonstrating safe performance in multi-agent and enterprise environments.
Applications and use cases
7
Digital Trust and Cyber Resilience in Internet of Things Ecosystems
Recent IoT ecosystems combine inexpensive devices, cloud platforms, vendors, and opaque data flows, making it difficult to establish responsibility and sustain trust after compromise. Research gaps include lifecycle assurance, device retirement, cross-vendor accountability, and user comprehension of security signals. This proposal could create a measurable trust-resilience model linking technical controls with perceived reliability. Objectives include mapping ecosystem dependencies, designing adaptive attestation, evaluating recovery behavior, and informing certification schemes for connected environments.
Applications and use cases
8
Fair and Explainable Digital Trust Mechanisms for Financial Technology Platforms
Trust in fintech is undermined when automated risk decisions are opaque, discriminatory, or difficult to challenge, particularly for underserved populations. Existing explainability methods rarely connect technical explanations with procedural fairness and consumer confidence. This study should design trust mechanisms that combine interpretable models, contestability, privacy safeguards, and independent oversight. Objectives include identifying harmful trust asymmetries, testing explanation designs, evaluating outcomes across demographic groups, and developing practical governance guidance for responsible financial automation.
Applications and use cases
9
Trust Recovery After Data Breaches Through Transparent Organizational Communication
Meanwhile, organizations continue to lose customer trust after breaches despite issuing standardized notifications that often appear defensive, delayed, or incomplete. The gap is limited evidence about which communication, remediation, and compensation practices genuinely restore trust across affected groups. This research should develop a process model of post-breach trust recovery and test interventions experimentally and longitudinally. Objectives include linking disclosure quality to behavioral outcomes, identifying credibility signals, and producing evidence-based crisis communication protocols.
Applications and use cases
10
Trusted Cross-Border Data Sharing Using Federated Learning and Data Spaces
Finally, cross-border data collaboration is constrained by privacy laws, incompatible governance, sovereignty concerns, and limited confidence in remote participants. Federated learning and data spaces promise cooperation without centralizing data, yet their trust assumptions, incentive structures, and failure responses remain underdeveloped. This project should design a governance and technical architecture for trustworthy collaboration. Objectives include modeling participant risks, improving privacy and robustness, evaluating incentive compatibility, and validating the approach in multinational sectoral pilots.
Applications and use cases