Chart 2 — H3 baseline vs recent distribution

The how. Where Chart 1 shows when the H3 line diverged, this chart shows how the H3 underwriting-risk-score distribution shape shifted. Mass moved right, into the decline / non-QM lane.

What you are looking at

• X-axis: underwriting-risk-score (0 to 100)
• Y-axis: density of applications at that score for H3
• Blue distribution: baseline window (first 30 days), what H3 scoring normally looks like
• Orange distribution: recent window (last 30 days), what H3 scoring looked like at Snapshot time
• Vertical line at score 67: threshold for routing to decline / non-QM lane

What the shape shift means

The orange distribution's mass has moved right compared to the blue distribution. Three things are happening at the same time:

1. The peak (mode) moves up. The most common H3 score used to sit in the manual-review band; now it sits at the edge of decline.
2. The right tail grows. More H3 applications score above the decline / non-QM threshold.
3. The left tail shrinks. Fewer H3 applications score in the auto-approve range.

What KL-divergence quantifies

Kullback-Leibler divergence measures how different one distribution is from another. Values:

• KL < 0.05: the same distribution (random noise)
• KL 0.05-0.2: mild shift, monitor
• KL 0.2-0.5: significant shape shift, flag as medium event
• KL 0.5-1.0: substantial shape shift, high-severity event
• KL > 1.0: distributions barely overlap — the population the AI is scoring is behaviorally different from the population it was trained on

The sample's H3 KL-divergence hit 0.833 on Day 44 and 0.963 on Day 65. Well above threshold, and it stayed there. H5 also fired on Days 30, 37, and 79.

The underwriting + regulatory read

The next chart

The baseline-vs-recent chart shows how the H3 distribution shifted. The lane-rate chart shows the operational consequence: which lane are H3 files getting routed to now vs baseline.

Chart 3 — lane rate per applicant group →