From login to lockdown: Inside a platform that stops fraud in real time

AI driven fraud is accelerating, with deepfakes and synthetic identities expected to drive $40 billion in losses by 2027. Brad Cranford, Director of Product Strategy at Alkami, shares how their digital banking platform delivers layered defenses, using advanced identity scoring, behavioral biometrics, and real time anomaly detection to block deepfakes, account takeovers, and synthetic identities before losses occur.

Don’t miss the other insights, strategies, and tools to help your institution stay ahead of evolving fraud threats in the Deep Dive: Fraud is everyone’s fight.

TRANSCRIPT: 

Brad Cranford, Alkami: 

Hello, my name is Brad Cranford. I’m a director of Product Strategy at Alkami Technology. We deliver digital banking solutions for banks and credit unions that deepen relationships, drive revenue, and most importantly, keep your institutions and users secure at every touchpoint. 

Fraud is evolving rapidly, and AI was powering the next wave of threats. That includes increasingly sophisticated tactics like deepfakes, which are now being used to impersonate real users, manipulate staff and bypass security systems.  

According to the Deloitte Center for Financial Services, AI driven fraud is expected to cause 40 billion in losses by 2027 with a 32% annual growth rate. That’s a steep curve, and deepfakes are a major contributor to that rise. 

Here’s the challenge. While one in 10 financial executives say they’ve already encountered a deepfake, 61% admit that they’re not prepared, they don’t have the processes and procedures in place to respond, detect, and be prepared for deepfake attacks and other AI tools. That’s a dangerous gap. 

At Alkami, we believe it’s critical that your digital banking platform isn’t just secure. It must actively help you detect and stop fraud in real-time from the first interaction to every transaction thereafter. 

Let’s walk through how we do this. Starting right at account opening. With our recent acquisition of MANTL, we’ve integrated an omnichannel account opening solution that blocks fraudulent actors before they can even get in the door. 

A major focus here is detecting synthetic identities where fraudsters piece together real data from multiple individuals to create fake but convincing personas. To combat that, we use advanced identity scoring and device and IP intelligence. This helps us spot patterns like repeated use of IPs or suspicious device behaviors that often indicate fraudulent attempts, including those using AI generated personas. This helps us stop bad actors from creating accounts that open doors for other activities within your platform like mule accounts at the very beginning, once an accountant is open, protecting ongoing interactions is just as critical, and especially as deepfakes become more realistic and harder to detect. 

Digital ID is one of our key tools for the digital banking platform in this area. It integrates in with your call center allowing your staff to verify callers using our mobile app. This ensures that the person on the phone is actually the verified user and not someone using a deepfake voice or pre-recorded audio to manipulate your staff. It also shields your employees from social engineering scams, which deepfakes increasingly enable and gives them the confidence that they’re talking to who they think they’re talking to when they’re helping your customers. 

We also use behavioral biometrics within the digital banking experience, analyzing how users naturally navigate your app to detect if something’s off. For example, if a bot or impersonator may move differently, hesitate in odd places or interact in ways that don’t match the user’s typical patterns. This includes detecting items that could show that a user is being manipulated by a bad actor on the phone, so it’s the real user using the device that someone’s talking them through, and so they’re behaving in an unusual pattern. 

We can detect this, raise that risk, and stop the action before it happens so the user doesn’t have losses, and you’re not chasing money after it leaves your platform. Deepfakes are often used in account takeover attacks where fraudsters attempt to move funds quickly and erratically. Our transaction anomaly detection tools identify when users perform actions that don’t align with their usual behavior, like transferring money to new accounts or unusual volumes or unusual times. These can trigger MFA challenges or automatically block suspicious activity again in real time, preventing funds from leaving your system so you’re not chasing those funds later. 

For your commercial customers, we also offer ACH and check positive pay. These tools help detect altered or unauthorized payments, giving your business clients a direct way to stop fraudulent transactions before they process, including check washing and other manipulations of check and ACH. This gives your customers confidence that the payments that they’re making are the ones they intend to make to real people and have not been manipulated. 

All of these tools are part of a broader platform designed to fight back against AI powered threats like deepfakes and continued ongoing evolution of financial fraud. Fraud is no longer just a risk, it’s a daily reality. Bad actors are using smarter tools and continuing to evolve.  

At Alkami, we’re committed to giving financial institutions the layered defenses they need to see fraud coming to stop it in real time. 

If you’d like to learn more, please visit us at alkami.com. Thank you very much.