Cybersecurity | AI Blurs Truth and Falsehood, Digital Identity Verification Must Be Strengthened!

During the 2025 International Fraud Awareness Week, cross-industry survey data from the Association of Certified Fraud Examiners (ACFE) and SAS sounded the alarm: as AI blurs the line between truth and falsehood, strengthening digital identity verification has become a critical defence.

The findings underscore the severity of deepfake social engineering attacks: 77% of respondents reported a sharp rise in such attacks over the past two years, while 83% anticipate moderate (28%) or significant (55%) growth in the next two years. The core reason lies in the criminal dividends yielded by Deepfake technology's enhanced ‘ease of use’ and ‘realism’. Fraudsters can rapidly generate false facial features, voices and other information via AI to impersonate relatives, friends, executives and other identities. By exploiting trust to breach defences, they substantially increase success rates.

AI simplifies fraud by blurring the line between truth and falsehood, rendering traditional identity verification ineffective. In conventional scenarios, victims might discern authenticity through vocal patterns or facial details. However, Deepfakes can precisely replicate a target's voice inflections, mannerisms, and facial expressions, rendering sensory-based verification impossible for the average person. For instance, fraudsters may use AI to mimic an executive's voice for transfer instructions or generate fake official videos to induce information disclosure. Such attacks are highly deceptive, posing significant financial and reputational risks to individuals and businesses.

To counter these new AI fraud challenges, strengthening digital identity verification is paramount. Single-factor authentication methods (such as passwords or SMS verification codes) have clear limitations, necessitating the development of a multi-dimensional intelligent verification system. Technologically, advanced biometric techniques like liveness detection, 3D facial recognition, and dynamic voiceprint verification can precisely distinguish genuine identities from AI-generated forgeries. Concurrently, big data analytics should be employed to study user behaviour patterns, establishing anomaly alert mechanisms to intercept risks proactively.

Beyond technological upgrades, industry collaboration and public education are equally vital. Enterprises must proactively enhance internal verification systems, implementing multiple layers of security for critical operations like financial transfers and permission changes. Relevant authorities should expedite the development of industry standards and regulations to govern technology usage and combat criminal exploitation of deepfakes. Concurrently, platforms such as International Fraud Awareness Week should be leveraged to educate the public about AI fraud, disseminate detection methods, and elevate nationwide vigilance.