BMT-05.04 Executive Summary#
BlueMirror.tech | May 2026#
Segment-based personalization produces recommendations for “78-year-old Black women in the Midwest.” The recommendation is a statistical average of that segment, which means it is wrong for every specific person in the segment. Margaret is a 78-year-old Black woman in Gary, Indiana, but she is also a former teacher, a churchgoer, a grandmother, an insomniac, a gardener, afraid of hospitals, and proud of her independence. These dimensions interact. Being Black and diabetic in Gary means something different from being white and diabetic in Palo Alto, because the healthcare infrastructure, the pharmacy access, the financial context, and the cultural framing of the diagnosis are all different.
The Intersectional Context Engine (I-ICE) rejects the categorical approach. Identity dimensions do not add. They multiply. The interaction between dimensions creates specific experiences that are not accessible through any single dimension alone. I-ICE captures identity across six dimension families: demographic, socioeconomic, health, social, psychological, and temporal, each containing multiple measurable attributes. The list is not fixed. If the person identifies a dimension the system has not encountered (“I am a veteran”), a new dimension is created with its own salience patterns.
Context-dependent salience determines which dimensions matter in each interaction. For a health query, race and ethnicity carry high salience because medication metabolism varies across populations. Religious affiliation carries low salience. For a social connection query, religious affiliation carries high salience because church community is primary social infrastructure for many aging adults. The salience weights are learned, not hard-coded. If Margaret’s religious affiliation turns out to be relevant to her health decisions, the system increases the salience of that dimension in health contexts for Margaret specifically. The population-level weights are a starting point. Her individual salience profile is the product.
The interaction effects between dimensions emerge from observation, not programming. The system learns through Margaret’s interactions that her independence orientation interacts with her hospital avoidance to produce a preference for home-based health monitoring, reinforced by her geography and financial constraints. A different person with the same demographic profile but different psychological dimensions would produce different recommendations.
The design constraint governing I-ICE is the line between personalization and stereotyping. The system never predicts preferences from dimensions alone. It observes preferences through behavior and uses dimensions to contextualize those observations. If Margaret’s preferences diverge from any population pattern associated with her dimensions, the system follows Margaret, not the pattern. Her data overrides her category. The privacy model extends per-dimension: Margaret may want her pharmacy to know she is diabetic but not her race. Each dimension has a visibility flag per external party.
The full article is available at BlueMirror.tech.