Where we’ve been and what’s next: Understanding fraud prevention in banking

What’s up fraud fighters, and welcome to Fraud Forward!

This episode is just me and you, and honestly, that felt right. Fraud Forward has always been about real conversations, but every once in a while, I think it matters to pause, zoom out, and talk peer to peer about where we’ve been, what I’ve been building, and where this work is going next. No guests. No panel. No polished back-and-forth. Just a real check-in on fraud prevention in banking and the decisions that are shaping what comes next.

This episode matters because the questions in front of us are getting bigger, not smaller. We are no longer just talking about fraud trends in isolation. We are talking about AI in banking, payment authorization, customer dispute resolution in banks, real-time decisioning, and the governance problems sitting underneath all of it. If you work in fraud, risk, payments, compliance, or digital banking, you already know this is not a future-state conversation. It is here now.

And that is really the core of this episode. Fraud prevention in banking is no longer just about identifying bad activity after the fact. It is about whether institutions can make better decisions earlier, with better context, better governance, and better alignment across teams. That is the shift. That is what I have been hearing in conference hallways, seeing in operator conversations, and building toward behind the scenes.

I also wanted to make this episode personal because that matters too. I have sat in the practitioner seat. I know what it feels like to defend fraud losses in a boardroom, to get asked how your institution compares to peers, to explain why a rule fired, and to carry the pressure of making the right call with incomplete information. So when I talk about AI agents in banking, banking fraud detection, or responsible AI in banking, I am not talking from a distance. I am talking from the perspective of somebody who knows what it feels like when the operational reality hits the fraud desk first.

Here is what that shift means in practice:

• We need to stop treating fraud as only a detection problem and start treating it as a governance and decisioning problem.
• We need better ways to evaluate customer authorization fraud, disputes, and real-time transaction risk before losses harden.
• We need clearer operator-level benchmarking so banks and credit unions can advocate for resources with facts, not guesswork.
• We need stronger collaboration across fraud, compliance, product, operations, and leadership because no one team can solve this alone.

What you’ll hear in this episode:

• A look back at the biggest themes from recent Fraud Forward conversations, including AI, KYC, payment fraud, and systems-level risk.
• What I have been building behind the scenes to support fraud fighters with clearer signals, faster context, and more practical guidance.
• Why benchmarking has become one of the most important gaps to solve for community banks and credit unions.
• What I am seeing across conferences, operator conversations, and industry events right now.
• My take on why scams, chargebacks, first-party fraud, and regulation are all colliding in ways the industry can no longer ignore.

You should listen to this episode if you:

• Work in fraud prevention in banking and need a clearer view of where the industry is heading
• Want a peer-level perspective on AI in banking, governance, and transaction risk
• Are trying to connect fraud operations to executive conversations about budget, staffing, and strategy
• Need stronger language and better framing around customer dispute resolution in banks and customer authorization fraud
• Care about building fraud programs that are more proactive, more defensible, and more useful to the institution

If you liked this episode, be sure to subscribe and review the podcast on iTunes, Spotify, YouTube, or wherever you listen to podcasts. It really helps with getting the word out.

Fraud prevention in banking is shifting from detection to decisioning

Banking is not just about spotting the bad transactions anymore. It is about how we make the call in real time, how we govern that call, and whether the right teams are actually looking at the right signals before it becomes a loss.

Because let’s be honest, a lot of what banks are dealing with right now does not fit neatly into the old model. AI agents in banking are starting to influence decisions. Payment authorization is getting more complicated. Disputes are getting messier because customer intent, manipulation, and first-party behavior are all running together. And somehow fraud teams are still expected to work these cases like it all starts and ends with one suspicious transaction. We know better than that.

That is why the real problem is not just banking fraud detection. It is whether the institution has the governance, the context, and the internal alignment to make the right decision sooner instead of cleaning it up later.

• Fraud is no longer just a detection problem. It is a governance and decisioning problem.
• AI in banking is forcing institutions to rethink how decisions are made and documented.
• Banking transactions need stronger context, not just faster alerts.
• Responsible AI in banking depends on whether institutions are ready to govern what those systems influence.

What I’ve been building and why it matters to fraud fighters

A big part of this episode is about the work happening behind the scenes and why it matters operationally. I wanted to share that because I think fraud fighters deserve more than polished messaging. You deserve to know whether the people building products and frameworks actually understand what it feels like to sit in your seat.

That is why this episode centers on practical clarity. Whether it was the NACHA webinar and blog focused on fraud monitoring updates, the internal work happening with banking product teams, or the Five Minutes of Fraud and Monday Fraud Fix formats, the goal has stayed the same: less noise, more signal.

The point is not to create more content for the sake of content. The point is to help fraud, compliance, and risk teams make faster, more informed decisions in environments where speed and clarity matter.

• AI governance in financial services has to be translated into operational language, not just policy language.
• Banking compliance and AI discussions are only useful when they connect directly to real workflows and real controls.
• Fraud fighters need concise, signal-rich context they can actually use in meetings, investigations, and strategy conversations.
• Building trust infrastructure in banking starts with listening to the people doing the work every day.

Why benchmarking is a missing layer in banking risk management

For years, fraud fighters have been asked to answer a question that sounds simple but is actually really hard. How do we compare to our peers? And the truth is, most of the time, we have not had a solid answer. Not because we were not asking the right questions, but because we did not have a credible, operator-built source of truth to point to. And that becomes a real problem when you are trying to justify a new analyst, explain fraud losses to leadership, or make the case for technology your team actually needs.

That is why benchmarking is not just a reporting exercise. It is part of banking risk management. Because if you do not know how your fraud losses, your staffing model, or your tools compare to institutions like yours, then you are walking into strategic conversations without the context you need. And that is a tough place to be when you are the one expected to have the answer.

That is exactly why I spent time in this episode talking about what this benchmarking effort is meant to solve. Not sponsored fluff. Not solution-driven bias. Real data built for operator decisions, because that is what fraud teams actually need.

• Community banks and credit unions need peer-based fraud benchmarking they can use in real strategic conversations.
• Better benchmarking supports staffing, technology investment, and risk assessment decisions.
• Fraud prevention in banking gets stronger when institutions can compare losses, controls, and program maturity with context.
• Operator-built benchmarking creates better decision support than vendor-shaped narratives.

Scams, chargebacks, and first-party fraud are colliding

Fraud teams used to be able to lean on one pretty simple question: did the customer authorize it? But today, that question by itself is not enough.

A scam victim may send the payment because they truly believe the fraudster is legitimate. A customer may dispute a transaction they knowingly made. Someone may file a chargeback on a purchase that was completely valid. On paper, those situations can look really similar. But they do not mean the same thing, and that is where things start to get messy.

And that matters for customer dispute resolution in banks, because the old frameworks were built for a world where the lines felt a lot clearer. Today, they are not. Fraud, abuse, manipulation, and customer intent are all running together. So if institutions want to make better decisions, they need stronger ways to step back, look at the full context, and understand what actually happened.

• Customer authorization fraud is creating new gray areas for fraud and dispute teams.
• Banks need to move beyond simple confirmation models and into behavioral and contextual analysis.
• Banking fraud detection must account for scams happening with the customer, not just to the customer.
• Transaction governance for banks has to evolve to handle intent, manipulation, and abuse more accurately.

Regulation still lags the operational reality

Fraud has changed fast. But a lot of the frameworks around liability, prevention, and accountability still have not caught up. And that is a real problem. Financial institutions are still being treated like the last stop for financial responsibility, even when the scam started somewhere else, picked up speed across other channels, and only touched the bank at the very end.

That puts fraud teams in a really tough spot. Expectations keep climbing. Loss tolerance keeps shrinking. But control over the full fraud chain is still limited. And honestly, that is one of the clearest reasons why fraud prevention in banking cannot be talked about in a silo anymore. It has to be part of a bigger conversation around AI, compliance, dispute resolution, and who is actually responsible across the full ecosystem.

Because the real issue is not whether fraud is shared. We already know it is. The real issue is whether the industry is finally ready to align responsibility with the point where prevention could have actually happened.

• Financial institutions are still absorbing losses tied to scams that begin outside the bank.
• AI and compliance for banks must be part of a broader conversation about shared responsibility.
• Payment authorization frameworks and customer reimbursement models are under pressure.
• Fraud teams are being asked to solve systemic problems without controlling all of the levers.

This episode is really about momentum. It is about where the conversation has been, what is finally getting clearer, and what still needs to be built. The industry does not need more vague awareness. It needs better signals, stronger collaboration, and more honest operator-level conversations about what fraud teams are actually facing. That is where this show is headed, and that is the standard I want Fraud Forward to keep pushing toward.