Michigan State University Federal Credit Union avoids $2.57 million in deepfake fraud exposure with Pindrop

Deepfake-driven fraud has become an increasingly acute issue for financial institutions, and Michigan State University Federal Credit Union wanted outside help detecting fraudulent deepfake calls in its call center. The $8.26 billion-asset credit union had three goals: reducing fraud without adding friction, increasing operational efficiency to reduce costs and improving the member side of the contact center experience. It also lacked a system with full visibility into fraud exposure in the contact center, although it had a good handle on transactional fraud monitoring.

The credit union partnered with Pindrop, an information security company that provides risk scoring for phone calls to detect fraud and authenticate callers, and launched its Passport and Protect products in the call center in August 2024. Pindrop's systems run a series of passive authentications on each call before the member is connected to an agent, replacing the manual security questions agents previously asked at the start of a call. The agent sees the risk assessment on their desktop and can adjust accordingly.

Between the launch and September 2025, the 367,000-member credit union traced $2.57 million in avoided fraud exposure from catching deepfake calls. Passive authentication cut an average of 58 seconds from authentication time per call, and the credit union's Net Promoter Score increased by 10% from 57 to 63 immediately after implementation. That period was the first year the credit union precisely tracked avoided fraud exposure in its call center.

• Passive call scoring before an agent picks up can strengthen security and shorten calls at the same time.
• Measuring avoided fraud exposure in the contact center gives leaders visibility that transactional monitoring alone does not.
• Credit unions and small banks are more likely to buy specialist fraud tools than build them, which makes vendor partnerships central to their programs.