Dashboard
AI Overview
AI Generated Summary
Cross-Model AI Health Brief
Across 247 production models, AI processed 18.4M predictions today at a blended 97.8% forecast accuracy. Revenue Forecasting v2.7 and Fraud Detection v4.1 are the strongest correlated performers — high volume, high accuracy, high business impact. One model, Churn Prediction v3.2, is drifting 2.1% below its accuracy baseline and is now flagged for review. Recommendation: prioritize retraining Churn Prediction v3.2 this week to protect its attributed $4.2M in business impact.
Explain This AI Summary
Model
Cross-Model Health Aggregator v1.6 (correlates volume, accuracy & business impact across all 247 production models)
Data Sources
Live telemetry from Churn Prediction v3.2, Revenue Forecasting v2.7, Fraud Detection v4.1, Customer Segmentation v1.9, Demand Planning v2.3 and Sentiment Analysis v1.4, synced within the last 5 minutes.
Why 96% confidence
Cross-validated against 14 days of prediction volume and accuracy telemetry with ±1.8% variance; all contributing models except one scored above the 90% minimum accuracy threshold.
Why Churn Prediction v3.2 was flagged
Its 7-day rolling accuracy fell from 96.9% to 94.8%, a 2.1% drift that exceeds the 1.5% alert threshold, while its prediction volume held steady — indicating a model quality issue rather than a data volume issue.
247
Models active
18.4M
Predictions today
99.98%
API availability
96.4
Business health
Model Performance Correlation
Bubble size = business impactX: prediction volume (M/day) · Y: accuracy (%)
Prediction Volume Trend
Total vs. top contributing models · last 14 days
Model Comparison
Top models by production traffic · click a card to drill down
Churn Prediction
v3.2
94.8%
Accuracy
3.2M
Volume/day
$4.2M
Impact
Revenue Forecasting
v2.7
97.1%
Accuracy
2.1M
Volume/day
$6.4M
Impact
Fraud Detection
v4.1
98.9%
Accuracy
4.6M
Volume/day
$3.8M
Impact
Customer Segmentation
v1.9
91.3%
Accuracy
1.4M
Volume/day
$2.1M
Impact
Demand Planning
v2.3
88.6%
Accuracy
1.8M
Volume/day
$1.9M
Impact
Sentiment Analysis
v1.4
82.4%
Accuracy
0.9M
Volume/day
$0.6M
Impact
No models match this filter.
Model Drift Detection
1 model driftingRolling accuracy drift · last 5 weeks
Model Lifecycle Status
247 production models by lifecycle stage
Goal Tracking Center
Scenario Simulator
Model a change in overall prediction volume
Projected Volume
18.4M
Projected Impact
$16.3M