Credit Card Fraud Model

A 98% accurate model with three critical vulnerabilities

How OrbTech audited a production-grade fraud detection model trained on 284,807 real transactions - and what it found beyond the accuracy score.

Dataset
Kaggle Fraud284,807 transactions
Model
Random Forestsklearn, depth 10
Scan time
Under 60s8 checks automated
Risk level
MEDIUM 42/1002 HIGH findings

The question accuracy can't answer

This case study documents an OrbTech security audit of a credit card fraud detection model trained on the Kaggle Credit Card Fraud dataset - one of the most widely used public benchmarks in financial ML. It represents the kind of production-grade classifier Indian fintech teams deploy every day: high accuracy, real transaction data, binary classification.

The audit set out to answer one question: does an excellent accuracy score mean a model is safe to deploy under DPDP Act, GDPR, and EU AI Act requirements?

The short answer: no.

All 8 checks, one scan

OrbTech ran every security check automatically in a single 60-second scan. Here is the full result set:

CheckResultSeverityRegulatory Relevance
Baseline PerformanceROC-AUC 0.9867, Acc 98.0%PASSEU AI Act Art.9
Feature PerturbationHigh accuracy dropHIGHEU AI Act Art.15
Boundary SearchMinimal exploitabilityMINIMALEU AI Act Art.15
Membership Inference73.3% inference accuracyMEDIUMGDPR Art.35, DPDP
Feature IntegrityNo over-reliancePASSEU AI Act Art.13
Data PoisoningNo anomaliesPASSEU AI Act Art.10
Model InversionModerate riskMEDIUMGDPR Art.5, Art.25
Model Stealing100% surrogate agreementHIGHIP / Trade Secret

Key findings explained

Model Stealing

HIGH

A surrogate model achieved 100% agreement with the original across test predictions. An attacker could reconstruct a functionally identical copy of the model purely by querying its prediction API - no access to training data or source code required. For a team that invested months of data science work, that is a direct IP and competitive risk.

Regulatory relevance: constitutes misappropriation of a proprietary AI system, triggering IP protection obligations under Indian IP law.

Membership Inference

MEDIUM

The scanner determined whether a specific record was in the training set with 73.3% accuracy (random chance is 50%). If the training data included real customer transactions, an attacker could probe the model to confirm whether a specific individual's data was used - a genuine privacy breach.

Regulatory relevance: GDPR Article 35 requires a DPIA for large-scale personal data processing. DPDP Act 2023 imposes similar obligations on Indian data fiduciaries. A 73.3% result is direct evidence a DPIA is required.

Feature Perturbation

HIGH

Under deliberate input perturbation - small, crafted noise added to transaction features - the model's fraud detection accuracy dropped sharply. This simulates exactly the manipulation a sophisticated fraudster would attempt: crafting transactions to evade detection without triggering obvious anomalies.

Regulatory relevance: EU AI Act Article 15 requires high-risk AI systems to withstand adversarial manipulation. A fintech fraud model is a strong candidate for high-risk classification under Annex III.

"A model that scores 98% accuracy can still leak private data, be stolen, and be fooled by adversarial inputs. Accuracy measures performance - it says nothing about security."

Read the complete audit

The full PDF includes model and dataset specs, all 8 check results, detailed regulatory mapping across GDPR, DPDP Act, EU AI Act and ISO 42001, and remediation steps for every finding.

↓ Download case study PDF ↓ See the actual scanner output Run it on your own model →

Built for Indian ML teams

OrbTech runs 8 automated security checks - adversarial attacks, privacy analysis, model theft simulation, data integrity - on any sklearn, XGBoost, LightGBM, or Keras model, in under 60 seconds. Every scan generates a PDF audit report with risk scores, plain-English findings for compliance teams, technical detail for engineers, and regulatory mapping to EU AI Act, GDPR, ISO 42001, and DPDP Act 2023.

Want to audit your own model? DM Shubham Kumar on LinkedIn or email security@orbtech.in for a free invite code.