1、0Using AI for Fraud DetectionPT Nomura Research Institute Indonesia2024 Special Report 1Only 2%of Fraud in Southeast Asia is Detected Through Data MonitoringIs AI the Key to Safeguarding Businesses from Fraud?The Answer is Yes.Financial fraud is rising worldwide as the adoption of advanced technolog

2、ies opens new avenues for online criminals,according to a recent INTERPOL report.Southeast Asian countries are also experiencing this trend,with many businesses falling victim to fraudsters.Yet,only 2%of companies in the region use technology to detect fraud.While AI can be misused for fraud,it also

3、 holds significant potential to protect businesses from becoming victims.222342850203483354341050100Southeast Asian Corruption Perception IndexAverage Score for Asia Pacific RegionThe Corruption Perceptions Index(CPI)*1 is the worlds most widely recognized ranking for measuring corruption.It assesse

4、s how corrupt the public and private sectors are perceived to be in each country,based on insights from experts and business leaders.Nearly 80%of Southeast Asian countries have a CPI score below the global average.In comparison,the Asia Pacific regions average score is 45 on a scale of 0 to 100,wher

5、e 0 indicates severe corruption and 100 represents a corruption-free environment.Fraud Cases in Southeast Asia(SEA)Corruption Is A Big Issue 3Occupational fraudfraud committed by individuals against their own employersis highly prevalent in Southeast Asia.Research from the Association of Certified F

6、raud Examiners(ACFE)reveals that the most common forms of occupational fraud in the region include false invoicing,fictitious expenses,and manipulated financial statements.*24The research further indicates that many companies have yet to adopt technology and data analytics for fraud detection.Tips r

7、emain the most common method for uncovering fraud,accounting for 48%of cases,while only 2%are detected through data monitoring.*2 This highlights a significant opportunity for AI technology to address this gap,particularly in the Southeast Asian region.48%16%13%4%4%4%4%2%2%2%2%1%TipManagement Review

8、Internal AuditExternal AuditDocument ExaminationBy AccidentAccount ReconciliationNotification by Law EnforcementAutomated Transaction/Data MonitoringConfessionOtherSurveillanceHow Is Occupational Fraud Initially Detected in the Asia-Pacific Region?5The Role of AI in Fraud DetectionWhat AI Can Do?Com

9、mon procurement fraud schemes include vendor collusion,bid rigging,bribery,and false invoicing.AI can play a critical role in combating these issues by analyzing vendor data from various sources,including legal documents,contracts,email correspondence,historical transactions,and other relevant data.

10、Through advanced algorithms and machine learning,AI can detect patterns and anomalies that may signal suspicious activity or potential collusion.This enables the early detection of fraud,helping organizations mitigate risks and maintain the integrity of their procurement processes.Procurement1.Fraud

11、 frequently occurs in claims and reimbursements involving forged documents or fictitious expenses.The manual process of verifying supporting documents can be both time-consuming and labor-intensive.AI offers a powerful solution by automating the analysis and verification of claim data.By scrutinizin

12、g each claim,AI can detect irregularities or suspicious documentation,automatically flagging them for further investigation.This streamlines the verification process and significantly improves the accuracy and efficiency of fraud detection.Claim&Reimbursement2.6Traditional fraud detection in account

13、ing typically relies on manual auditing,a process that is time-consuming and prone to human error.In contrast,AI can quickly analyze vast amounts of financial data with greater accuracy.By training AI to establish a baseline of normal financial activities,it can detect patterns and identify deviatio

14、ns in real-time.Incoming transactions are continuously monitored and compared against these established baselines,with any irregularities flagged for further investigation by fraud analysts.This approach significantly improves both speed and accuracy while reducing the risk of oversight.Accounting3.

15、Phishing,malware attacks,and hacking are some of the most common threats to identity verification.AI can play a key role in preventing identity theft and mitigating unauthorized transactions caused by scams.For example,AI models can be trained to cross-check data across multiple platforms or channel

16、s to verify the authenticity of a transaction and detect abnormal patterns that may signal fraud.A real-world example is Singtels solution,SingVerify,a network-based authentication system designed to prevent fraud.*3 Unlike traditional methods such as MFA or 2FA,SingVerify operates in real-time with

17、out human intervention,verifying identities by matching users device data with telco network data in the background.Transactions are automatically denied if the data doesnt match.This AI-driven process reduces fraud risks while providing users with enhanced security and a seamless experience.Securit

18、y&Identity Verification4.7The Role of AI in Fraud DetectionCommon Challenges in AI ImplementationImplementing new technological solutions,especially AI,is rarely straightforward.It often comes with challenges and risks,ranging from technical issues and human factors to external variables beyond the

19、organizations control.Recognizing these potential hurdles early on is crucial for successful AI adoption.8Implementing new technological solutions,especially AI,is rarely a straightforward process.It often comes with various challenges and risks,ranging from technical issues and human factors to ext

20、ernal variables beyond the organizations control.Recognizing these potential hurdles early on is crucial for successful AI adoption.Data Quality&Availability:AIs effectiveness relies on high-quality,well-structured data.Without this,AI capabilities cannot be fully leveraged.Integration with Legacy S

21、ystems:Incorporating AI into existing enterprise systems can be difficult due to compatibility issues with legacy infrastructure.Thorough planning is required to ensure a smooth integration.Data Privacy:AI systems process vast amounts of data,including sensitive or personally identifiable informatio

22、n.This raises serious concerns about data privacy and the potential for breaches.Ethical Concerns:If not carefully managed,the data used to train AI models could introduce bias,leading to stereotyping or discrimination in decision-making processes.Regulatory Compliance:The regulatory framework for A

23、I is still evolving,particularly in Southeast Asia.While the ASEAN organization has developed the ASEAN Guide on AI Governance and Ethics*4,AI policy maturity varies across the region,with most countries in the early stages of AI adoption.9The Role of AI in Fraud ManagementThe Role of AI in Fraud De

24、tectionFind the Right AI for Your Business With thousands of AI solutions on the market,choosing the most suitable one for your organization can be challenging.To make an informed decision,consider the following factors:By considering these factors,organizations can narrow down the options and choos

25、e an AI solution that is not only effective but also aligns with the business needs and financial resources.10SummaryFraud and corruption threaten businesses,leading to financial losses,reputational damage,and legal consequences.Adopting AI for fraud detection is a critical step toward addressing th

26、ese issues effectively.AI brings powerful capabilities,such as precise and efficient transaction analysis,behavior-based analytics,anomaly detection,and real-time alerts for high-priority issues.By harnessing AI technology,organizations can proactively monitor risks and minimize financial damage cau

27、sed by fraud and corruption.However,implementing AI for fraud detection comes with its own set of challenges.Despite these hurdles,the benefits of AI-enhanced detection,faster response,and greater accuracy make it a worthwhile investment for organizations committed to safeguarding their assets and r

28、eputation.AuthorRiska SalamIT System Consultant,Nomura Research Institute Indonesia Contactnri-syscon-linkedinnri.co.jp Learn More About Our IT System Consulting ServicesReferences1 Corruption Perception Index:www.transparency.org/cpi2 Occupational Fraud 2024:A Report to the Nations:https:/ What is SingVerify?:https:/ ASEAN Guide on AI Governance and Ethics:https:/asean.org/wp-content/uploads/2024/02/ASEAN-Guide-on-AI-Governance-and-Ethics_beautified_201223_v2.pdf2024 Nomura Research Institute,Ltd.All Rights Reserved.