Originally published on CUInsight.com
At the recent financial industry event, one question dominated every conversation: How do we stop our members from losing everything to scams? For credit unions serving older, less tech-savvy communities, this isn’t an abstract risk management discussion. It’s a crisis unfolding in real time at teller windows across the country.
The answer isn’t what most fraud officers expect. It isn’t better fraud detection systems or more sophisticated transaction monitoring. The answer is preventing the fraud before it starts. Because by the time your 80-year-old member is standing at the wire transfer window, convinced her grandson needs bail money, you’ve already lost.
The numbers tell a devastating story
The scale of elder fraud has reached crisis proportions. In 2024, Americans over 60 lost $4.885 billion to scams, a 46% increase from 2023. The Federal Trade Commission’s data is even more alarming: fraud losses reported by older adults increased about fourfold from 2020 to 2024, skyrocketing from $600 million to $2.4 billion.
But these numbers only capture what gets reported. The FTC estimates the real losses experienced by older adults in 2024 may be as much as $81.5 billion. When older adults reported losing more than $100,000, the number of reports increased nearly sevenfold, and combined losses went up eightfold.
From the 2024 Consumer Sentinel Network: Adults in their 70s reported losing a median of $1,000 to scams, compared with $417 for those in their 20s. For investment scams specifically, those in their 70s lost a median of $20,000, compared to $1,551 for victims in their 20s.
The AI threat: When your voice can be weaponized
If traditional scams represent a crisis, AI-driven fraud represents an existential threat. Modern AI can clone a person’s voice with 85% accuracy using just 3-5 seconds of audio, enough for a Facebook video, a voicemail, or a Zoom call saying “Happy Birthday, Grandma.”
According to Resemble AI’s Q1 2025 Deepfake Incident Report, global losses from deepfake-enabled fraud reached over $200 million in Q1 2025 alone. Deloitte’s Center for Financial Services projects that AI-enabled fraud losses will reach $40 billion by 2027 in the United States, up from $12 billion in 2023. A Sumsub report found that in the Asia-Pacific region, AI-related fraud attempts surged 194% in 2024. A McAfee survey revealed that 70% of people aren’t confident they can tell the difference between a cloned voice and a real one.
Scammers increasingly direct victims toward irreversible payments. According to the FTC’s 2024 data, 33% of older adults who reported losing $10,000 or more to imposter scams indicated cryptocurrency was the method of payment. Your 75-year-old member who’s never owned Bitcoin is being coached through a Bitcoin ATM by someone she believes is a government agent. Fewer than 5% of these funds are ever recovered.
Why detection alone fails
By the time fraud detection systems fire, members are already emotionally invested. They’ve been on the phone with scammers for hours. They believe they’re saving their grandson or protecting their money. When you call to verify, they may lie because they think you’re the problem.
Traditional detection looks for unusual patterns. But when your member genuinely believes they’re helping family, there’s no anomaly. AI-driven scams exploit urgency, authority, emotion, and familiarity.
Credit unions cannot out-technology international criminal syndicates. But you can out-educate them.
Education before victimization
Building a security & fraud center
Randolph-Brooks Federal Credit Union’s Security & Fraud Center provides members a direct way to verify suspicious communications. Members submit screenshots of texts, emails, or messages, and RBFCU identifies whether they’re scams.
Older members often sense something’s wrong but aren’t sure. They need immediate verification from a trusted source. Implementation should include:
- Direct submission portal for uploading screenshots
- Same-day or next-day verification
- Clear yes/no answers with next steps
- Library of real scam examples
In-person workshops
Monthly or quarterly workshops build critical defenses:
- Pattern recognition: Show real scam scenarios. Play actual recordings. Let members hear the tactics. Pattern recognition requires repetition.
- Demonstrate AI: Play voice cloning examples. Many don’t believe it’s convincing until they hear it. Understanding the threat raises their guard.
- Verification protocols: Teach family code words. Role-play emergency calls. Make it concrete: “Ask them the name of their first pet. If they can’t answer, it’s a scam.”
- Cryptocurrency reality: “If anyone tells you to use a Bitcoin ATM to protect your money, it is 100% a scam. No legitimate organization will ever say this.”
Webinars and community engagement
Regular webinars, both recorded and on-demand, extend reach. Partner with law enforcement for credibility. FBI agents presenting current trends carry weight.
Develop “Protecting Your Parents” guides for adult children covering warning signs and how to discuss scams without condescension. Host multi-generational workshops where families create verification protocols together.
Partner with senior centers and community groups to deliver education where seniors gather. Provide presentation materials and train volunteers.
Prevention-focused education empowers members. Instead of potential victims, they become informed consumers capable of self-protection. Every dollar lost to scammers drains wealth from your community. Prevention is both member service and community development.
The AI arms race reality
AI-driven scams will worsen. Deepfake video calls will become indistinguishable from real ones. Future deepfakes will replicate emotional responses and personality traits.
But education creates defenses technology cannot bypass. The member who’s practiced verification will use it automatically, even when distraught. The member who knows voices can be cloned will pause before sending money. The member who understands Bitcoin ATM demands signal scams will hang up.
The question isn’t whether your older members will be targeted. They will be. The question is whether, when that call comes, when that voice sounds exactly like their grandson, they’ll remember to ask for the code word. Whether they’ll pause long enough to think: “This feels wrong.”
That pause can only be created through education. And supporting that pause requires more than awareness campaigns. It requires operational readiness, clear workflows, and teams equipped to respond quickly when prevention fails.
If your credit union is evaluating how to strengthen both prevention and response, now is the time to assess where gaps exist across education, intake, investigation, and recovery. The institutions that will protect their members best are those aligning proactive education with modern dispute operations.