AI-driven fraud prevention is transforming online trust, precision, and real-time decision-making.
Doriel, you’ve had a remarkable career as a Principal Technologist at Forter, a leader in e-commerce trust and fraud prevention. Can you share a bit about your journey and what drew you to focus on fraud prevention and identity protection in the digital space?
Interestingly, my professional journey began in Biology until I came across a Facebook listing for a role “catching online thieves.” I was intrigued by the post and applied to the position; that’s how I joined Forter over 10 years ago. Starting as a fraud analyst, I reviewed transactions across industries—from apparel to gift cards—which gave me unique insights into the dynamics of e-commerce fraud. Over time, I transitioned into roles supporting analytics and risk in the U.S., which deepened my understanding of the challenges faced by fraud professionals globally. The human aspect of this field has been what’s kept me passionate about it. Fraud prevention isn’t just about algorithms but also about understanding the people on both sides of the transaction. It’s one of the few career paths where the requirement is to think like a fraudster.
AI has become an integral part of fraud prevention in e-commerce. How do you see AI transforming the way fraud is identified and prevented in real time?
AI has significantly transformed fraud detection and prevention in e-commerce by enabling real-time, scalable responses to fraudulent activity. The key advantage of AI lies in its ability to process vast amounts of data quickly, detecting patterns and anomalies that would be impossible for humans to spot at such speed. AI’s ability to analyze transaction histories, recognize suspicious behaviors, and make decisions based on real-time data enables businesses to respond instantaneously, preventing fraud before it even impacts the consumer or business. This dynamic detection mechanism provides a faster, more proactive approach than traditional fraud prevention tools, which often rely on slower, reactive methods. Moreover, AI agents are becoming essential in identifying and distinguishing fraudulent patterns that evolve rapidly, keeping businesses one step ahead of increasingly sophisticated fraudsters.
However, AI isn’t just about speed; precision is also an important capability since identifying fraudulent users or behavior isn’t always black and white. For example, some online shoppers might appear suspicious at first glance but are legitimate upon closer inspection. AI is only as smart as its human operators, and the biggest analytic challenge at Forter is “teaching” the AI to think and act like an expert fraud analyst. This ensures that legitimate customers aren’t penalized while bad actors are caught.
Agentic AI seems to be a key player in reshaping e-commerce fraud strategies. How do you define agentic AI, and in what ways is it enhancing the ability of online businesses to protect consumers?
Agentic AI refers to AI systems that can act independently to accomplish specific tasks, such as making purchase decisions or navigating online platforms. While this offers incredible convenience for consumers, it also presents new challenges for fraud prevention. Distinguishing between a “bad bot” exploiting vulnerabilities and a “good agent” acting on behalf of a legitimate user has become critical. The key is focusing on understanding the identity and intent behind these interactions, not just reviewing the transaction as a standalone. This means enabling businesses to embrace the benefits of AI agents while staying ahead of malicious actors.
Fraud patterns often vary by region. How do different growth rates and regional economic factors impact fraud behaviors, and what should global merchants know about these dynamics?
Fraud patterns are heavily shaped by regional economic factors, growth rates, and local market conditions. Economic shifts like changes in tariffs, taxes, or supply chain disruptions can create new opportunities for fraudsters to target specific products or markets. In regions with increased product scarcity or price fluctuations, fraudsters may focus on exploiting high-demand items, such as electronics or high-value goods, where detection can be more challenging.
Regional behaviors, like frequent gift card use, can also reflect cultural or economic trends in certain localities. Fraudsters often target these transactions due to their anonymity and lack of tracking, making them an attractive target for fraud.
As another example, consumers across different regions may have varying reasons for shipping items to addresses other than their billing address – such as sending gifts, securing deliveries in safer locations, or managing multiple residences due to work or family. These behaviors can be culturally influenced, but it’s essential for businesses to differentiate between legitimate regional practices and potential fraud. Understanding these nuances is key for businesses to maintain trust with good customers while effectively safeguarding against fraud.
With the rising transaction volumes in the digital economy, the issue of chargebacks is becoming a growing concern. What strategies do you recommend for businesses to mitigate chargebacks effectively?
A key strategy is focusing on the “who” behind a transaction, not just the “what.” This involves understanding the context of the purchase — such as why a customer might use a different shipping address — so fraud detection systems don’t unfairly penalize legitimate users. Chargebacks will always be issued; an improved customer verification, clear documentation of why a transaction was approved, and AI-powered analysis of transaction behavior can all help in disputing chargebacks that were issued under false pretences. Additionally, investing in near real-time fraud prevention helps businesses stop fraudulent transactions before they happen, further reducing chargeback volume.
As AI continues to evolve, how do you see it influencing not just fraud prevention, but also customer experience and transaction speed in the e-commerce industry?
AI is transforming the customer experience by making transactions faster and smoother. For instance, it enables businesses to evaluate transactions in real time, reducing friction and ensuring customers can complete their purchases seamlessly. Moreover, AI helps personalize experiences by identifying trusted users and offering VIP-like treatment, such as loyalty perks or expedited checkouts. When implemented well, AI-based tools don’t just prevent fraud; they build trust and loyalty by ensuring every interaction feels secure and effortless.
Fraud prevention often involves striking a balance between security and customer convenience. How does Forter approach this challenge, and how do you ensure seamless user experiences while maintaining robust fraud protection?
At Forter, our focus is on combining AI’s strengths with subject matter experts’ judgment to adapt to constantly evolving fraud tactics. Striking a balance between fraud protection and customer convenience is essential. Fraud detection systems need to operate behind the scenes, so legitimate customers don’t experience friction during the checkout process, yet bad actors can’t go through. In most cases, catching the fraud is easier than identifying when a good user is mistakenly being flagged as a bad one. It all boils down to who is making the transaction rather than what the transaction is – by focusing on identifying the good users in real time based on their online behavior and historical context, we can enable that seamless experience and at the same time stop fraudulent activity.
As transaction volumes surge, the risks related to account takeover (ATO) increase. What steps are most effective in preventing ATO, and how is AI helping in identifying suspicious account activities early on?
Preventing ATO amid an influx of digital transaction volumes requires a combination of monitoring and proactive measures. AI plays a key role in detecting suspicious activities early by analyzing patterns in login attempts, transaction behavior, and device usage. For instance, if an account is accessed from an unusual location or device, AI can flag these behaviors and request additional verification before completing the transaction. Additionally, businesses should implement multi-factor authentication (MFA) and real-time alerts to notify customers of any unusual activity. Similarly to a transaction, the goal is to understand who the user is that is interacting with the account and whether they are the actual account owner. By using AI to monitor account activity continuously, businesses can catch signs of ATO before fraudsters have a chance to fully take over an account.
The role of real-time data networks in fraud detection is often emphasized. How do these networks integrate with AI to deliver immediate responses to fraud attempts and what makes them so effective in today’s landscape?
Real-time data networks are critical for detecting fraud as soon as it happens, and AI makes them even more effective by processing data quickly and accurately. By integrating AI with real-time data networks, businesses can instantly assess the legitimacy of a transaction based on a variety of factors, such as transaction history, device information, and geographic location. AI systems can also prioritize transactions that show signs of suspicious activity, enabling fraud prevention teams to focus on high-risk transactions first. These networks are effective because they allow businesses to act immediately to block fraud without introducing delays or creating friction for legitimate customers.
Finally, looking ahead, how do you envision the future of fraud prevention in e-commerce? What upcoming trends or technologies do you think will have the greatest impact on how businesses protect both their operations and their customers?
AI and Generative AI are tools that are used by frauds as well, and it’s important to always remember that. Any fraud prevention system that doesn’t leverage AI today is already missing the mark. The future of fraud prevention will likely feature even more advanced uses of AI and automation to detect and prevent fraud before it occurs. AI will become even more adept at recognizing subtle fraud patterns as it evolves on an ongoing basis, making fraud detection both faster and more accurate. The proliferation of AI agents will also be crucial for both fraud detection and customer experience management.
QUOTE : One quote I keep coming back to is something Steve Jobs once said, and I think holds water to this day and resonates with my answers above: “Technology is nothing. What’s important is that you have faith in people, that they’re basically good and smart, and if you give them tools, they’ll do wonderful things with them. Tools are just tools. They either work, or they don’t work.”
Doriel Abrahams
Principal Technologist at Forter
Doriel Abrahams is the Principal Technologist at Forter — the Trust Platform for digital commerce — where he monitors emerging trends in the fight against fraudsters, including new fraud rings, attacker MOs, rising technologies, etc. His mission is to provide digital commerce leaders with the latest risk intel so they can adapt and get ahead of what’s to come.