Green Sheet interviews ParaScript's Emiliano Giacchetti

As check fraud enters a more dangerous and complex phase, Emiliano Giacchetti, CEO of ParaScript, points to a perfect storm: the rise of sophisticated, tech-driven schemes and the growing average value of individual checks. No longer the domain of lone forgers, today’s fraud is fueled by organized groups wielding advanced tools capable of outpacing traditional defenses. In this Q&A, Giacchetti explains how artificial intelligence is reshaping fraud detection, why financial institutions must rethink outdated risk models, and what it takes to stay ahead in an era where every high-value check is a potential target.

1. You've described today as a "new era of check fraud." What are the significant differences between today's check fraud landscape and what FIs faced just five years ago?

Check fraud has evolved significantly in the past five years. What used to be mostly individuals forging checks on their own has evolved into coordinated efforts by organized groups using sophisticated tools and tactics. Fraudsters are quick to adopt new technology, often outpacing the defenses financial institutions put in place.

What makes this new era of check fraud particularly dangerous is that while fewer checks are being written, the value of each check is rising, and fraud is following the money, leading to greater losses for financial institutions and consumers.

According to the Federal Reserve, check volume fell to 11.2 billion in 2021, but the total value soared to $27.23 trillion. And, in 2023 alone, 65 percent of U.S. organizations reported experiencing check fraud, with related Suspicious Activity Reports nearly doubling over the past two years.

No payment channel is immune, but checks are now the second most targeted payment method after debit cards. We've seen a sharp rise in certain types of check fraud, especially mail theft leading to washed checks. The increased use of remote and mobile deposits has also added new risks, like duplicate deposits.

2. With fewer checks in circulation but higher average values, what are the implications for FI risk models and resource allocation?

As the average value of checks rises, a single fraudulent transaction can cause significant financial damage, especially with multiple ways to deposit or cash it. Because of this, financial institutions need to shift from volume-based risk models to value-based analysis.

This means prioritizing the transaction amount, doing a thorough analysis of the client profile, and using a combination of modern prevention and detection tools like check wash detection, signature verification or endorsement reading tools, to verify the integrity and origin of each check.

Choosing the right technology is crucial but challenging, given the many options on the market. Financial institutions should focus on AI-based solutions with proven accuracy and strong reputations rather than generic tools that aren't fine-tuned for check fraud and can produce unreliable results.

3. How AI-driven fraud detection systems able to identify and stop threats like advanced check washing or synthetic identity fraud in real time?

Traditional fraud detection systems struggle with real-time verification, especially at ATMs. For example, signature verification requires a reference signature, which banks store when an account is opened. While this verification may occur during in-branch transactions, it doesn't happen at ATMs, where there's no access to signature databases.

As a result, forged checks can be accepted without detection, and by the time fraud is uncovered, the funds are often already in the fraudster's account. ParaScript's proprietary, highly accurate and scalable AI-powered software addresses this by secretly embedding a reference signature directly onto the check in a secure, machine-readable format, enabling real-time verification even at ATMs.

Another example is check washing, where fraudsters alter parts of a legitimate check - usually the payee name or amount – while leaving the rest intact. Our technology analyzes handwriting across different zones of the check to detect inconsistencies. If the handwriting on the payee or amount line doesn't match the writing style used in other fields, the system flags it in real time. This allows financial institutions to catch altered checks immediately, even without access to historical handwriting samples.

In addition to signature and handwriting analysis, our software can also evaluate behavioral patterns such as signing speed, stroke order, and the sequence in which a check was completed, when reference data is available. This biometric analysis adds another layer of verification, helping detect inconsistencies that may signal fraud.

4. One key challenge you mentioned is balancing risk assessment with operational efficiency. How can FIs avoid slowing down legitimate transactions while remaining vigilant against fraud?

To fight fraud without slowing down legitimate transactions, financial institutions need AI-powered software. Modern platforms like ParaScript can process over 25 million checks daily, reduce fraud by up to 90 percent, and verify high-value items in real time, all while cutting processing costs by as much as 95 percent.

With self-learning algorithms and real-time updates, these systems adapt quickly to new threats. They also minimize false positives and negatives, helping financial institutions make smarter decisions aligned with their own risk policies and ensuring strong protection without added friction.

5. What strategies do you recommend for FIs to not only meet compliance requirements but also demonstrate proactive fraud mitigation efforts?

As regulatory scrutiny grows, financial institutions should adopt a risk-based approach to fraud prevention focusing on high-value, high-risk areas like debit card, check and mobile deposit fraud. Assessing risks across multiple channels and customer segments ensures resources are used efficiently and effectively.

Moreover, technology must be applied thoughtfully with clear, explainable, and auditable results, as blindly adopting AI without transparency and strategy can lead to compliance risks. Strong escalation protocols, centralized fraud data, and thorough reporting are also vital to prevent repeat fraud and demonstrate proactive management.

6. Looking ahead, how do you see fraud tactics evolving—and what should FIs be doing now to stay ahead of the next wave of threats?

Predicting exactly how fraud tactics will evolve is challenging because fraudsters are often more creative than those trying to stop them. What is clear, is that fraud will continue to grow in complexity, fueled by advanced tools that enable new scams and increasingly sophisticated approaches.

Detection will become harder as fraud becomes more personalized; for example, we're already seeing AI-generated avatars that blur the line between real and fake identities.

To combat this, the best path forward is building holistic and adaptive fraud-prevention strategies powered by AI that unify detection across all channels – be it mobile, teller, ATM, remote deposit, online and back office. The ability to capture and analyze data in real time is also crucial, because delayed detection often means the loss has already occurred.

Moreover, preventive tools should be prioritized alongside detection ones, ensuring financial institutions not only respond to fraud but stay ahead in stopping it before the damage happens.

Notice to readers: These are archived articles. Contact information, links and other details may be out of date. We regret any inconvenience.