PAYMENTS & COMPLIANCE


From student startup to fraud-fi ghting powerhouse: How SEON’s co-founder is rewriting the rules of risk


When fraud attacks nearly destroyed a student-built crypto exchange, SEON co-founder Tamas Kadar decided the industry’s tools were broken. In this interview, he explains why transparency, real-time AI and unifi ed risk systems are reshaping fraud prevention, and how companies can turn compliance from a burden into a strategic advantage.


GIO: What personal experience or industry gap ultimately led you to co-found SEON? Tamas Kadar: SEON wasn’t a planned venture. My co-founder Bence and I were building a cryptocurrency exchange when fraud attacks nearly brought the whole business down. As Bence put it at the time, the company was “pretty much going to burn to the ground.”


When we looked for help, the available tools were slow, opaque and created huge friction for legitimate users while trying to block fraudsters. That trade-off never made sense to us. We weren’t cybersecurity experts; we were university students facing a problem we couldn’t fi nd a good solution for. So, we built one ourselves.


GIO: What problem were you most determined to solve from day one? TK: The false choice between security and growth. Businesses were often told they had to accept either fraud losses or a painful customer experience. We believed that was a failure of imagination, not an unavoidable reality. Our aim was to stop fraud without blocking legitimate customers, and to do it transparently rather than through a ‘black box.’ Risk teams should be able to understand why decisions are made, explain them to regulators and stand behind them. That idea was unusual in the industry at the time, but it still guides our product decisions today.


GIO: Fraud tactics evolve constantly. How has the threat landscape changed in recent years? TK: The biggest shift has been the industrialisation of fraud. Generative AI has dramatically lowered the barrier to entry. Synthetic identities can now be assembled in minutes; phishing attacks are hyper-personalised at scale and the groups behind them operate like sophisticated businesses. The attack cycle has also compressed – from weeks to days, sometimes hours. Many organisations still rely on risk frameworks designed for a much slower era, which leaves them exposed. In sectors such as iGaming, bonus abuse and multi-accounting remain persistent problems because the effort required to create fraudulent identities is low compared to the reward. The operators pulling ahead are those moving away from reactive, rules-based systems toward real-time


12 MARCH 2026 GIO


platforms that detect patterns across networks rather than individual cases.


GIO: AI is central to both committing and preventing fraud. How do you ensure systems remain accurate and trustworthy? TK: Explainability was a core design principle for us from the beginning. Fraud teams must understand why a decision was made and be able to defend it internally and to regulators. A score that simply says “reject” without context isn’t useful.


Our approach combines machine learning with a rules engine that analysts can inspect and adjust. AI adapts to emerging patterns, but human judgment remains central. The goal is not to replace fraud teams but to give them better information to make decisions. Supervised machine learning is also key because fraudsters constantly change tactics. Systems must keep learning and evolving or they quickly fall behind.


GIO: Why is a unifi ed approach to fraud, AML and compliance becoming essential? TK: Fraudsters don’t operate in silos. A single bad actor might trigger signals across fraud detection, anti-money-laundering checks and identity verifi cation simultaneously. If those teams operate on separate systems and datasets, each only sees part of the picture.


There’s also a growing operational cost to fragmentation. Multiple vendors, integrations and review queues slow decisions and increase manual work. As regulations tighten globally, that ineffi ciency becomes harder to justify. The organisations pulling ahead are consolidating these functions into a single platform where fraud signals, AML screening and identity checks inform each other in real time. We call this a ‘command centre’ approach because it refl ects how modern risk operations need to function.


GIO: SEON has grown alongside highly regulated industries. How can compliance become a competitive advantage? TK: Compliance becomes a cost centre when it’s treated as a box-ticking exercise. When it’s embedded into the product from the start, it becomes something very different.


Transparency and explainability – two principles central to SEON – are exactly what regulators want. Our customers don’t have to retrofi t audit trails or explainable decisions because those capabilities already exist.


Speed also matters. Regulations in markets like the UK, EU and across APAC are evolving quickly. Businesses that can demonstrate real-time monitoring and clear decision logic are better positioned during licensing processes and regulatory reviews. In some cases, clients even use their fraud and AML capabilities as a selling point with their own customers.


GIO: Is there a customer outcome that captures SEON’s real-world impact? TK: The cases that stay with me are those where the stakes go beyond revenue. One client identifi ed a coordinated fraud ring and spent six months working alongside law enforcement to bring the perpetrators to justice. Their team later received formal recognition from the authorities. That’s a reminder that fraud prevention protects real people, not just balance sheets. At scale, the results matter too. We’ve helped companies reduce fraudulent account creation by up to 90% and signifi cantly cut manual review workloads. For industries like iGaming – where bonus abuse, payment fraud and AML obligations intersect – a unifi ed platform can completely transform how risk teams operate.


GIO: What will defi ne best-in-class fraud prevention over the next fi ve years? TK: Three things: speed, automation and unifi cation. First, systems must adapt in real time. Fraud cycles are shrinking, and platforms that update weekly or monthly won’t keep up. Second, fraud operations will shift from dashboards to more automated workfl ows. AI systems will connect signals across data sources and surface the most important information so analysts can focus on judgment rather than data gathering.


Finally, fraud, AML and identity verifi cation will increasingly run from a single command centre. Companies that adopt this model early will gain advantages in cost, effi ciency and regulatory readiness.


Page 1  |  Page 2  |  Page 3  |  Page 4  |  Page 5  |  Page 6  |  Page 7  |  Page 8  |  Page 9  |  Page 10  |  Page 11  |  Page 12  |  Page 13  |  Page 14  |  Page 15  |  Page 16  |  Page 17  |  Page 18  |  Page 19  |  Page 20  |  Page 21  |  Page 22  |  Page 23  |  Page 24  |  Page 25  |  Page 26  |  Page 27  |  Page 28  |  Page 29  |  Page 30  |  Page 31  |  Page 32  |  Page 33  |  Page 34  |  Page 35  |  Page 36  |  Page 37  |  Page 38  |  Page 39  |  Page 40  |  Page 41  |  Page 42  |  Page 43  |  Page 44  |  Page 45  |  Page 46  |  Page 47  |  Page 48  |  Page 49  |  Page 50  |  Page 51  |  Page 52  |  Page 53  |  Page 54  |  Page 55  |  Page 56  |  Page 57  |  Page 58  |  Page 59  |  Page 60  |  Page 61  |  Page 62  |  Page 63  |  Page 64  |  Page 65  |  Page 66  |  Page 67  |  Page 68  |  Page 69  |  Page 70  |  Page 71  |  Page 72  |  Page 73  |  Page 74  |  Page 75  |  Page 76  |  Page 77  |  Page 78  |  Page 79  |  Page 80