Tackling the Growing Threat of Deepfakes in Financial Services
The rise of deepfake technology breeds a new cyber threat for financial institutions. As the tech behind these realistic fake videos and audio files becomes more advanced, it’s increasingly being used for fraudulent activities, identity theft, and damaging misinformation.
Salinawati Salehuddin, Section Head of IT Risk Management at Agrobank, speaks with Corinium’s Eleen Meleng about the growing risks of deepfakes, how AI can help defend against them, and why maintaining customer trust is critical.
The Growing Risk of Deepfakes in Financial Services
Salehuddin pointed out that deepfakes are being used in more dangerous ways than ever before. Cybercriminals are no longer just interested in stealing data or money—they're also using these fake videos and audio clips to blackmail, manipulate identities, and deceive people in ways that many aren’t prepared for.
"We’re already seeing cases of criminals using deepfakes to impersonate well-known figures or even someone’s family member to trick people into giving away personal details or money," she explained.
She also mentioned that deepfake technology could soon be used to create synthetic biometric data, like fake fingerprints or facial features, to bypass security measures such as facial recognition. This presents a major risk for financial institutions that rely heavily on biometric security.
Another area of concern is business espionage and social engineering attacks, where deepfakes could be used to manipulate internal communications or impersonate executives, leading to significant financial and reputational damage.
Using AI to Combat Deepfakes
As deepfakes become more sophisticated, Salehuddin emphasized that AI-driven solutions are critical in detecting and stopping these types of attacks. At Agrobank, AI algorithms are being developed to identify fake audio and video by looking for anomalies or inconsistencies. The bank is also working on real-time monitoring of social media and news outlets to spot and address any potential deepfake content early.
“We’re investing in AI tools that can detect deepfakes by analyzing patterns and behaviors that humans might miss,” Salehuddin said. She also highlighted that user behavior analysis could help identify unusual activities, such as changes in communication patterns, that might signal the use of a deepfake.
Though they’re still exploring cost-effective ways to deploy these technologies, Salehuddin noted that collaboration with external cybersecurity partners and internal fraud teams is crucial in building a strong defense against these emerging threats.
Preserving Public Trust in the Digital Age
One of the biggest challenges deepfakes present is their potential to erode public trust in digital content. With the rise of disinformation and fake news, keeping customers confident in the bank’s digital platforms is more important than ever. Salehuddin emphasized that transparency and open communication are essential in dealing with deepfakes and any other cyber threats.
“It’s important that customers know what’s happening. Being transparent, especially when there’s a potential threat, helps maintain their trust,” she explained.
She also stressed the need for educating customers on how to identify fake content and avoid falling victim to scams. At Agrobank, they use webinars, newsletters, and training programs to help customers understand what to look out for. The bank has also implemented two-factor authentication and real-time alerts for suspicious activity, providing an additional layer of security for customers.
Preparing for the Future
As deepfakes continue to evolve, Salehuddin urged financial institutions to stay proactive and ensure that their security measures keep up with the changing threat landscape. By fostering a culture of responsibility and collaboration, banks can better protect themselves and their customers from deepfake-related risks.
"We can’t afford to be reactive. We need to have evidence-based strategies and strong governance frameworks to help us deal with these emerging threats," she said.
In her final thoughts, Salehuddin emphasised that the key to combating deepfakes lies in AI-driven cybersecurity solutions, collaborative efforts, and continuous education—both for employees and customers. As deepfakes become an increasing concern, the financial industry must stay one step ahead to ensure trust and safety in the digital world.
To explore more about the latest trends of cybersecurity in Malaysia, we invite you to join us at CISO Malaysia 2025, happening on 20 February. If you are interested in speaking at the events, feel free to reach out to Eleen Meleng.