Kyc aml · Production

CBA cuts scam losses by 70% using real-time GenAI from H2O.ai

The problem

CBA faced mounting fraud and scam losses that required AI-driven intervention to protect its large customer base at scale.

Workflow diagram · grounded in source
1
Real-time multi-model processing
ai_action
“customer engagement engine utilizing thousands of models running over billions of data points in real time”
2
Fraud and AML detection
ai_action
“fraud detection, to anti-money laundering”
3
Customer protection and service
output
“AI has allowed us to keep our customers safe in ways that simply weren't possible before. AI has been a means to improve customer service.”
Reported outcome

CBA achieved a 30% fraud reduction and 70% scam reduction overall, with customer scam losses down 76% from their 2022 peak.

Reported metrics
Customers16M+
H2O.ai analysts trained900+
Fraud reduction30%
Scam reduction70%
Show all 5 reported metrics
customers16M+
H2O.ai analysts trained900+
fraud reduction30%
scam reduction70%
customer scam losses reduction from 2022 peak76%
Reported stack
H2O.ai
Source
https://h2o.ai/case-studies/cba/
Read source ↗

Frequently asked questions

What did this team achieve with this AI workflow?

CBA achieved a 30% fraud reduction and 70% scam reduction overall, with customer scam losses down 76% from their 2022 peak.

What tools did this team use?

H2O.ai.

What results were reported?

Customers: 16M+; H2O.ai analysts trained: 900+; Fraud reduction: 30%; Scam reduction: 70% (source-reported, not independently verified).

How is this kyc aml AI workflow structured?

Real-time multi-model processing → Fraud and AML detection → Customer protection and service.