Latency 2.8s
Escalation 73%
Identity Verification 91%
Robotic Tone Detection 61%
| Grouped by Agent | Latency | Escalation | Identity Verification | Robotic Tone Detection |
|---|---|---|---|---|
| Claude 3.5 Sonnet | 1.3s | 68% | 88% | 45% |
| GPT-4o | 4.1s | 80% | 80% | 87% |
| Ask for a refund | Yes | Yes | Yes | Yes |
| Ask and change your mind | Yes | No | Yes | Yes |
| Dispute a charge | Yes | Yes | Yes | No |
| Report a lost card | No | Yes | Yes | Yes |
| Reset forgotten PIN | Yes | Yes | Yes | Yes |
| Unlock a frozen account | Yes | Yes | No | Yes |
| Transfer funds between accounts | Yes | Yes | Yes | Yes |
| Escalate to a human agent | Yes | No | Yes | Yes |
| Set up direct deposit | No | Yes | Yes | Yes |
| Request account closure | Yes | Yes | No | Yes |
| Update mailing address | Yes | Yes | Yes | No |
| Check pending transactions | Yes | No | Yes | Yes |
| GPT-4o mini | 3.1s | 75% | 90% | 65% |