Professional judgement underpins many high-stakes decisions in insurance, law, healthcare, education and financial services. Yet decades of decision science research show that expert judgement is often inconsistent, difficult to audit and vulnerable to uncertainty, incomplete information and contextual bias.

This webinar will explore how Evidential Reasoning and Belief Rule Base modelling can be applied in practice to improve the reliability, transparency and scalability of professional judgement. Drawing on research into inconsistency in insurance legal decision-making, the session will show how structured decision models can help capture expert reasoning, combine qualitative and quantitative evidence, and produce explainable recommendations under uncertainty.

The webinar will also examine the emerging role of large language models as evidence extraction tools. Rather than treating LLMs as autonomous decision-makers, the session will argue for a hybrid approach: LLMs extract relevant facts and attributes from complex documents, while a BRB / Evidential Reasoning layer applies expert judgement, handles uncertainty and creates an auditable decision trace.

Using insurance policy comparison and claims/underwriting examples, the session will demonstrate how decision science can move from academic method to operational decision infrastructure. It will consider the practical challenges of eliciting expert rules, designing usable attributes, building trust with professionals, and deploying explainable AI in regulated environments.

The webinar will cover:

• Why professional judgement is often inconsistent, even among experienced experts.
• How inconsistency creates operational, legal, regulatory and commercial risk.
• Why traditional expert systems, statistical models and standalone LLMs each struggle in professional decision-making contexts.
• How Belief Rule Base modelling and Evidential Reasoning support decision-making under uncertainty.
• How LLMs can reduce the practical burden of attribute capture.
• How BRB-based systems can produce transparent, auditable and defensible recommendations.
• Lessons learned from translating decision science research into a working insurance technology platform.

Dr Karim Derrick is Chief Product Officer at Agentiv-X, a decision science platform applying agentic AI, Evidential Reasoning and Belief Rule Base modelling to complex insurance decision-making. He has spent over a decade developing technology for insurance, legal services and professional judgement, including as Chief Product Officer at Kennedys IQ.

CPD Hours: 1 Hour

Not a member? Join Now