Fraud Is a Social Act — And That Changes Everything

Fraud isn’t just a data anomaly. It’s a behavior, rooted in context.

For decades, detection systems have been trained to look for deviations in transaction data, outliers in amounts, timing, geography, or device usage. But that approach assumes fraud happens in isolation.

It doesn’t.

Fraud is a social act. It exploits pressure, opportunity, ambiguity, and, most importantly, vulnerable systems designed by people for people.

The numbers back this up:
According to the ACFE 2024 Report to the Nations, 85% of occupational fraud cases involved efforts to conceal the fraud through behaviors that weren’t necessarily abnormal in isolation, but revealed patterns when viewed through a social or organizational lens.

Why Traditional Detection Falls Short

Legacy detection flags:

• A sudden login from an unfamiliar location
  
  
• An unusually large transfer
  
  
• An abrupt change in device or channel
  
  

But these are symptoms. Not root causes.

What most systems miss are the conditions:

• Repeated high-friction touchpoints before a rule-breaking event
  
  
• Transaction clusters that follow recognizable “crisis” behavior timelines
  
  
• Patterns in user behavior that align with fraud risk models, like Cressey’s Fraud Triangle or the Crowe ABC model (Action, Behavior, Concealment)
  
  

In other words, they see the event, not the ecosystem.

Fortza’s Approach: Modeling Risk Contexts

At Fortza, we don’t monitor what users say. We don’t mine their communications.

What we do is model how users behave under specific psychosocial conditions (stress, desperation, opportunity windows) using behavioral risk modeling backed by decades of research.

Our detection strategy draws from:

• Cressey’s Fraud Triangle (1953): Pressure, Opportunity, Rationalization
  
  
• Crowe ABC framework (2019): Action, Behavior, Concealment
  
  
• Behavioral signals of deception (Vrij, 2008): Frequency, delay, and escalation patterns
  
  
• Research from the Association of Certified Fraud Examiners and the Behavioral Insights Team (UK)
  
  

We analyze:

• The sequence and escalation of actions
  
  
• The timing and repetition of high-risk behaviors
  
  
• Contextual anomalies within defined behavioral journeys
  
  
• Cross-platform signals that reveal environmental instability
  
  

These patterns tell us more than a single spike ever could. They help us predict fraud, not just react to it.

The Future of Detection is Context-Aware

Fraud is growing more sophisticated because it understands systems, not just software systems, but human ones.

Fortza is designed to answer:

• What behavior preceded this transaction?
  
  
• How does this fit into the user’s typical psychological context?
  
  
• Are we seeing the digital version of desperation, coercion, or manipulation?
  
  

These aren’t “gut feeling” questions anymore. They can be modeled, and we’re doing it.

Up Next

Next week, I’ll explore the difference between pattern recognition and motivation modeling, and why the systems that only focus on anomalies are chasing ghosts.

It’s time we stopped hunting outliers and started understanding why they happen.

References in comments.

#FraudDetection #BehavioralAnalytics #Cybersecurity #Fortza #PsychosocialAI #RiskContext

References

• Cressey, D. R. (1953). Other People's Money: A Study in the Social Psychology of Embezzlement.
  
  
• Vrij, A. (2008). Detecting Lies and Deceit: Pitfalls and Opportunities.
  
  
• Association of Certified Fraud Examiners (2024). Report to the Nations on Occupational Fraud and Abuse.
  
  
• Crowe LLP (2019). The Crowe Fraud Risk Management Guide: ABC Model.
  
  
• The Behavioral Insights Team (2021). Using Behavioral Science to Understand and Prevent Fraud.
  
  
• Verizon (2023). Data Breach Investigations Report.
  
  
• McKinsey & Company (2022). Digital Trust: How to Build It and Why It Matters.