Skip to main content
I, Cobot: Humans and Machines Team Up to Fight Fraud

It is a new buzzword for me: cobot, or collaborative robot, “a robot intended for direct human-robot interaction within a shared space.” And I got a few HAL vibes when reading a Harvard Business Review article on cobots that says:

“Firms achieve the most significant performance improvements when humans and machines work together. Through such collaborative intelligence, humans and AI actively enhance each other’s complementary strengths: the leadership, teamwork, creativity, and social skills of the former, and the speed, scalability, and quantitative capabilities of the latter.”

Well, as it happens, FICO is a leader in cobot technologies — we are at the forefront of using human-machine collaboration to fight fraud.

FICO CCS: A Cobot Pioneer

If you’ve ever received a fraud alert on your smart phone, you’re likely to be interacting with a cobot. FICO® Customer Communication Services (CCS) for Fraud uses two-way, multichannel communication to actively engage customers in the protection of their accounts, sending an automated, digital fraud alert that empowers them to resolve or identify suspected fraud on their accounts in real-time, at the point of sale.

Deceptively simple, CCS for Fraud gets humans and machines collaborating in a shared space: SMS, email, interactive voice or their bank’s mobile app. This communications solution leverages data and predictive analytics from FICO® Falcon® Fraud Manager to identify potentially fraudulent transactions, and then reaches out to customers via their preferred channel. By confirming whether the transaction in question is legitimate or not, the customer directs the machine to take the next step, e.g., to authorize or decline the transaction.

CCS for Fraud is widely used because this cobot is highly effective. FICO research shows that banks and card issuers using CCS for Fraud have realized 95% customer satisfaction, reduced declined transactions by 30%, and increased resolved fraud cases by 250%, with no additional overhead. Furthermore, by automating digital contact strategy, CCS for Fraud delivers operational improvements such as:

  • Immediate identification of suspicious activity and faster customer contact
  • Deeper penetration of queue volumes
  • Automating work that is performed manually, such as alert reviews and calls to the customer 

  • Increasing revenues by converting unnecessarily referred or declined transactions into approvals.

Next-Level Coboting: Robotic Process Automation

Humans and machines also work side-by-side in financial crimes case management, through FICO’s Financial Crime Solutions. FICO® Alert & Case Manager consolidates alerting and case management, using robotic process automation (RPA) to prioritize alert handling and investigation.

In this cobot instance, there’s an interplay between AI models, rules and case management. It starts with FICO Alert & Case Manager monitoring and consolidating alerts and cases from multiple source systems. Rules and analytics eliminate many manual steps, and determine when human intervention is needed, escalating cases to an investigator. When the likelihood of fraud is difficult to determine, the machine triages cases so investigators aren’t overwhelmed with false positives. Instead, human efforts can be focused on investigating and resolving the most important cases.

How AI-Driven Alert Prioritization Works

Diving deeper into that explanation, an AI-driven analytics score rank-orders alerts and provides a clear hierarchy for prioritizing investigative resources. These tasks are high-volume, repetitive and time-consuming, so they are prime candidates to be executed by robots. RPA ensures that processes and decisions are consistent. It also reduces errors, increases accuracy, and allows fraud management processes to operate autonomously 24/7.

As an example of how powerful cobots can be, an International Tier 1 Financial Institution (and FICO Alert & Case Manager customer) found that using RPA resulted in 95% of the alerts being autoworked within its service level agreement (SLA). In addition, the weekly screening hit rate was lowered by an average of 67% across regions.

Follow me on Twitter @FraudBird for more my thoughts on cobots, robots and the latest developments in payments and fighting fraud.

related posts