Fraud Forward

What’s up fraud fighters, and welcome back to Fraud Forward!This episode comes to you live from Money 2020 Las Vegas, where I sat down with four people whose work I admire. Angela Diaz from TD, Jared Tennison from GBank, Jeni Branter from Royal Credit Union, and Aaron Frye from Lucid Point Consulting. Together they cover bank risk oversight, quantitative fraud analytics, credit union payments and digital banking, and fraud risk consulting.I also brought along our new fraud benchmark report for credit unions and community banks, and I used its findings to shape my questions. We talk about who owns fraud, why loss numbers rarely match from one team to the next, how scams reach members in digital banking, and why many teams shop for a new tool before they understand the one they already have. If you have been wondering what good fraud governance looks like in practice, this conversation is for you. Let’s get into it.What you’ll hear in this episode:How Angela Diaz carried lessons from a startup into large financial institution work, and what makes cross functional fraud collaboration work or failWhat good fraud policy ownership looks like after the policy is written, including feedback from the people who have to apply itWhy Jared Tennison says fraud teams can give three different answers about losses, and why first party fraud is the hardest to pin downHow the hidden cost of fraud friction, including customers who never finish an application, rarely shows up in loss reportingWhich scams Jeni Branter sees reaching members in digital banking, including romance scams and bank impersonation callsWhy Jeni asks business members to meet security requirements in their contractsHow intentional friction, step up checks, and limits based on the member’s risk can protect members without slowing everyone downWhat Angela says changes when AI agents start to transact, and why she believes risk avoidance is not risk managementHow Aaron Frye explains the second line of defense, fraud risk appetite, and why fraud ownership keeps moving from department to departmentPractical steps before buying a tool, including building requirements, inventorying what your current tools cover, and asking about fourth party riskYou should listen to this episode if you:Lead or support a fraud program at a credit union or community bank and want practical ideas for governanceAre trying to explain why your loss numbers differ between teams or channelsAre preparing to replace a fraud vendor and want to know what to do firstWork at a smaller institution and want to build strong fraud governance with limited resourcesAre thinking about how AI agents will change accountability for transactionsWant more fraud fighting in your inbox? Subscribe to The Monday Fraud Fix. My weekly newsletter with the trends, stories, and takeaways I’m paying attention to. Subscribe here.Connect with Angela Diaz | LinkedInSenior Risk Manager, External Fraud Risk Oversight at TDConnect with Jared Tennison | LinkedInAVP Quantitative Risk and Fraud Team Lead at GBankConnect with Jeni Branter | LinkedInSVP Payments & Digital Services at Royal Credit UnionConnect with Aaron Frye | LinkedInFounder & CEO at Lucid Point ConsultingConnect with Hailey Windham, CFCS | LinkedIn 

What’s up fraud fighters, and welcome back to Fraud Forward!I first came across Assaf Kipnis through his work online, and more recently he shared a post about realizing he was more than an investigator. Assaf has spent 13 years in security, trust and safety, and fraud, with investigation roles at LinkedIn, Facebook, Google, and ElevenLabs. Today he runs his own consultancy focused on investigations, AI builds, and threat intelligence. In our conversation, he raises a question I hadn’t considered. While we as an industry are debating whether AI will replace fraud investigators, the people committing fraud are using AI to experiment, innovate, and improve their operations. What happens if we remove the very people who help us stay ahead?We talk about leveraging AI for fraud investigations in a way investigators can actually trust. That includes the AI fraud investigation tools that are being built, why context matters more than hallucinations, what human in the loop really means, and how a fraud knowledge base can keep institutional knowledge from walking out the door. Let’s get into it.What you’ll hear in this episode:Why modern fraud operations run like companies, with goals, leaders, and recruiting, and why taking down a thousand accounts doesn’t slow themWhy replacing fraud investigators with AI could become a competitive disadvantage for fraud teamsHow Assaf built an open source investigation tool that connects OSINT sources and saves evidence to diskWhy hallucinations are less of a risk than AI making decisions with incomplete or outdated contextWhat human in the loop and human on the loop mean in practice, and why reviewing every AI decision doesn’t scaleWhy most of an AI build should be deterministic code, and where a language model earns its placeHow an investigator’s adversarial mindset helps when building with AI, including treating the model like a threat actorHow a fraud knowledge base captures what experienced investigators know and keeps it after they leaveWhat fraud leaders should do instead of bolting AI onto existing processes You should listen to this episode if you:Wonder whether AI will replace fraud investigators, or have been asked to automate investigationsLead a fraud team and want a practical path for an AI transformation that keeps your expertiseAre an investigator or analyst who feels capped by not writing code and want to see what changes with AIWant to learn how to build AI fraud investigation tools with guardrails and traceable evidenceWorry about what happens to your team’s knowledge when experienced people leave

What’s up fraud fighters, and welcome back to Fraud Forward!We spend a lot of time talking about how scams work. This time we’re talking about how a scam feels while it’s happening to someone, and why one credit union decide that was the most important story to tell.Amanda Balmer, Director of Financial Crimes at SPECU, has joined me for conversations at conferences before. This is the first time she gets an episode of her own. Amanda leads a team of analysts, investigators, and strategists who protect members. And she’s a big part of why PSECU Fraud Files exists. It’s a video series the credit union built with its marketing team that shares real member stories, anonymized, to help other members spot the warning signs before they lose money.We dig into why PSECU moved from a newsletter to storytelling while protecting member privacy and dignity, all to build a series that is changing the way members and employees talk about fraud. We also look at the scams hitting credit unions right now, including how a fraudster can coach someone to lie to their own financial institution. If you have ever wondered how a story can treat member education as both an awareness effort and a fraud control, this one is for you.What you’ll hear in this episode: How Amanda went from criminal justice student to social worker, private investigator, intelligence analyst, and now director of financial crimesHow a team of analysts, investigators, and strategists works across PSECU to protect members and the credit unionWhy impostor scams, including bank impersonation scams and one-time passcode scams, are the trend having the biggest impact right nowHow fraudsters coach victims to get around the questions financial institutions askWhy PSECU treats member fraud education as both awareness and a fraud controlHow the Fraud Files evolved from a newsletter into fraud awareness videos built around real member storiesHow PSECU approaches anonymizing member fraud stories and takes the shame out of themA complete job scam story involving a fake check and a mobile deposit, and the scam red flags it revealsWhat frontline employees can notice in an ordinary conversation, and how these stories have changed internal training and empathy You should listen to this episode if you:Work in credit union fraud prevention and want a model for member education that goes beyond warning listsAre thinking about sharing real member stories and need a plan for privacy, marketing, and legal approvalLead or train frontline staff and want practical ways to bring fraud prevention into everyday conversationsWant to understand how job scams, fake checks, and mobile deposit check fraud look from the member’s point of viewSupport scam victims and want language that replaces blame with understandingAre trying to build a better working relationship with your marketing team

What’s up fraud fighters, and welcome back to Fraud Forward!I want to talk about something that might be a little uncomfortable. It isn’t about a new scheme or a criminal using AI to break into a bank. It’s about our own fraud controls, and the gap between having them and knowing they work. Two stories from the past week made me want to have this conversation. The first is a $350 million penalty against American Express National Bank tied to deficiencies in its BSA and AML program. The second is a guilty plea from a former TD Bank employee who admitted to accepting bribes and helping launder more than $4.8 million. The facts are very different, and neither institution looks like a community bank or credit union with a fraud team of two. Even so, the same questions show up at every size. Does your risk assessment reflect the business you run today? Who reviews the people with the power to override a control? And when someone finds a weakness, who makes sure it gets fixed?I’m also bringing in what we’ve learned from our financial institution fraud benchmarking research, because it points to a question I keep returning to. What does a mature fraud program actually look like? Let’s get into it.What you’ll hear in this episode:Why the American Express enforcement action is a lesson in fraud risk assessment, and what happens when an assessment focuses on a narrow part of a much larger businessWhy the roughly $13 billion in suspected activity cited by the OCC reflects deficiencies in a BSA and AML program, and why it shouldn’t be repeated as a fraud lossHow a former TD Bank employee allegedly used legitimate access to unblock restricted cards, and why employee override monitoring deserves more attention than it getsWhat 86 accounts at one commercial address teach us about transaction monitoring and what should happen after an alert firesPractical ways smaller teams can use compensating controls when they can’t fully separate dutiesWhy documenting a known control weakness is not the same as managing it, and how remediation tracking can quietly stallA simple framework that separates control existence vs execution vs effectivenessMetrics that show real exposure to leadership, beyond alert counts and cases closedFive questions to bring to your fraud team on MondayYou should listen to this episode if you:Lead or support a fraud program at a community bank or credit union and have to cover a lot of ground with a small teamWant to know whether your fraud risk assessment is driving decisions or only documenting themAre responsible for managing insider risk or employee access and want practical ways to review overridesNeed help explaining known control weaknesses to leadership in a way that leads to actionAre comparing your program against peers and want to know what a mature fraud program looks like

What's up fraud fighters and welcome back to Fraud Forward!I just got home from Las Vegas at four in the morning on Friday after starting the week in Washington DC. I am home for exactly two more sleeps before I turn around and leave again. Today there is no guest. It is just me, my microphone, and a notebook full of things to talk about.I started the week at HHS then I flew from DC to Vegas for ACAMS The Assembly. Two completely different rooms. Yet I kept writing down versions of the same thing. Culture. Data. Accountability. Collaboration. Speed. Technology.And then one sentence from Washington ended up framing the entire week for me. Fraud follows money. That is how this episode starts and that is how it ends. Where the money goes, fraud follows. When the money changes, fraud changes. When the way we move money gets faster, fraud gets faster. When we create a new product, somebody figures out how to exploit it. When we close one vulnerability, they find another.That is fraud driven flows in practice. And if you pay attention to where the money is going next, you will have a much better picture of where the fraud is going.What you’ll hear in this episode:Why fraud driven flows connects real-time payments fraud, stablecoin fraud risks, and check fraud in the digital age into one through lineGovernment fraud program banking parallels that reframe how we think about audit accountability and corrective actionWhy fraud prevention stewardship and the line a dollar saved is worth more than a dollar recovered should change how we talk to leadershipHow fraud program culture functions as a control and why fraud incentive misalignment quietly defeats everything else we buildWhat ACAMS Assembly 2026 insights on real-time payments fraud and stablecoin financial crime risks mean for fraud risk beyond organizational boundaryWhy fraud technology before problem definition is still one of the most expensive mistakes in fraud programs and what AI fraud results measurement should actually look likeHow fraud program single point of failure is a fraud risk hiding inside a staffing conversationWhy fraud collaboration operationalized looks nothing like fraud collaboration discussedYou should listen to this episode if you:Lead a fraud program and have ever felt like you are having the same conversation in completely different roomsAre trying to figure out how to measure fraud prevention value and tell a better story to leadershipCare about fraud program culture and want to hear why it functions as a control, not a soft conceptHave ever bought technology before understanding the problem you were trying to solveAre dealing with fraud knowledge retention risk because one person on your team knows everything and you are not sure what happens when they leave