RESEARCH PROJECTS
AI, Cyber Threat Intelligence, and decision-making under uncertainty
Applied research focused on developing models, frameworks, and protocols for analyzing uncertainty, Cyber Threat Intelligence, and decision support in digital scenarios involving significant exposure.
Alongside her published work, F.B. is currently developing a second monograph on synthetic uncertainty, Cyber Threat Intelligence, and digital evidence in democratic contexts involving significant exposure. The work also draws on a recent period of personal and professional experience in the Middle East during a time of war and regional instability. This experience is approached not as autobiography but as a situated vantage point: a concrete perspective from which to examine the convergence of physical infrastructure, digital services, institutional communications, cyber risk, crisis narratives, and the difficulty of rapidly verifying what is happening. The monograph begins with a central question: what happens when a technical, informational, or geopolitical event must be understood, documented, characterized, and communicated before all the evidence is available? Within this framework, the concept of synthetic uncertainty is developed as a legal-forensic category with a cyber intelligence focus: not generic uncertainty produced by the digital age, but an increase in the extent to which authenticity, provenance, integrity, attribution, or evidentiary value can be contested within a defined decision window. The volume connects this category with Cyber Threat Intelligence, digital forensics, public incident communications, the management of attribution, and election crisis scenarios leading up to the 2028 U.S. presidential election. Part of the research uses Dubai as a situated vantage point on the crisis of verifiability.
The aim is not to turn the experience in the Gulf into generalizable proof or to suggest that regional dynamics automatically recur in other settings. It is to identify mechanisms that may transfer only under certain conditions, such as causal ambiguity, compressed verification timelines, dependence on digital infrastructure, contestable attribution, and overlap between a physical event and a cyber narrative. The final part of the volume develops scenarios and tabletop exercises as stress-testing tools, not predictions. These scenarios test whether, when faced with leaks, synthetic content, impersonation, false claims, indirect disruptions, CTI poisoning, false flags, or post-election challenges to evidence, a system can still distinguish an event, a signal, a data point, an indicator, intelligence, evidence, legal proof, attribution, a decision, and communication. The value of these scenarios lies not in predicting the future but in making failure modes visible: overstated assurance, delayed communication, loss of evidence, circular reporting, automation bias, confusion between technical attribution and legal responsibility, and the transformation of narrative into apparent proof. Two methodological working papers are also underway on Cyber Exposure Decision Intelligence, with particular attention to vulnerability management, exposure management, the pre-CVE phase, and the quality of decision-making under uncertainty. The first paper formalizes a model of counterfactual remediation and verifiable residual risk, while the second paper defines a protocol for empirical validation and falsification of the model. These ongoing works are a natural continuation of my research on digital evidence, Cyber Threat Intelligence, AI-amplified risk, infrastructure vulnerability, democratic security, and decision governance. Their common thread is the same question that runs through my professional and academic work: how to produce reliable assessments when evidence is incomplete, threats are accelerating, attribution remains contestable, and a decision cannot wait for absolute certainty.
