We asked Viston to modernize our fraud detection and AML monitoring with explainable AI and graph analytics while meeting regulator expectations.

Business challenge

• High false positives in card fraud, rising mule activity, and manual KYC refresh.
  
• Fragmented rule engines and limited case investigation tooling.
  

Our approach and solution

• Built a streaming fraud scoring pipeline with graph features and explainable models.
  
• Upgraded AML monitoring with entity resolution and risk scoring; automated KYC refresh.
  

AI applications and benefits delivered

• Real-time fraud scoring with LightGBM and graph-derived features.
  
• Neo4j-based network risk scores to surface mule rings.
  
• SHAP-based explanations integrated into case management.
  
• LLM assistant for investigator summarization and SAR drafting with strict guardrails.
  

Cost of implementation

• USD 1,75,000 including data platform uplift and model validation.
  

Time to implement

• 7 months to full production across cards and retail payments.
  

Tools and technologies used

• Azure Databricks, Spark Structured Streaming, Kafka, Snowflake, Neo4j, LightGBM, SHAP, Elastic Stack, AKS, Key Vault, MLflow, Great Expectations, Power BI.
  

Quantitative outcomes

• 48% reduction in fraud losses on targeted products.
  
• 37% reduction in false positives.
  
• 60% faster KYC refresh turnaround.
  
• 22% higher investigator throughput.
  

Key performance indicators (KPIs) tracked

• Fraud loss rate bps, false positive rate, precision/recall, time-to-detect, AML alert productivity, SAR timeliness, model drift.
  

Pre- and post-implementation metrics

• Fraud loss rate: 7.8 bps → 4.1 bps.
  
• False positive rate: 92% → 58%.
  
• Average KYC TAT: 10 days → 4 days.
  
• Alerts per investigator per day: 18 → 22.
  

Stakeholder quotes or testimonials

• “Explainability made model sign-off straightforward with internal audit and the regulator.” — Head of Financial Crime.
  
• “Graph features exposed mule networks we couldn’t see with rules only.” — Fraud Strategy Lead.
  

Regulatory or compliance considerations

• UAE Central Bank AML/CFT guidance, FATF-aligned controls, PCI DSS scope containment, model risk policy adherence, challenger model testing, comprehensive audit logs, privacy-by-design.