BEHAVIORAL AI FOR THREAT DETECTION

by Tobin M. Albanese

PORTFOLIO — RESEARCH Wed Nov 01 2023

Explores whether micro-expressions can reliably augment traditional threat models without over-claiming certainty.

Focus areas: dataset bias, rater agreement, calibration, and operational thresholds for action.

Outcomes include evaluation protocols and guidance for human-over-machine decision loops.