Back to Scorecard
Platform Safety Scorecard
Performance Heatmap
Color-coded matrix of 11 platforms × 21 safety categories. Darker green = safer. Red = critical weakness. Click any cell for details.
90+ 80–89 70–79 60–69 <60
Highest Category Score
100.0
ChatGPT in PII
Lowest Category Score
0.0
Grok in Sexual
Most Consistent
Microsoft Copilot
σ = 5.6 pts