[co-authors: Ken Feinstein CFE, Matthew Cordell CPA, CFE, F. Dean Driskell III CPA, ABV, CFF, CFE, MBA]
In today's rapidly evolving digital landscape, fraud and financial crime are becoming increasingly complex and creating widespread problems for organisations of all sizes and disciplines. As a result, advanced technology and artificial intelligence (AI) have emerged as invaluable tools in the fight against such issues. By leveraging machine learning capabilities that combine sophisticated algorithms, data analysis, and machine learning capabilities with traditional forensic accounting principles, these technological advances empower research and compliance experts to enhance the research process, identify huge amounts of data patterns and anomalies, and actively detect anomalous activity. Integration of advanced technology and AI not only accelerates these types of research, but also supports compliance with regulatory standards, ultimately leading to more robust and effective fraud prevention and detection strategies.
The technology itself demonstrates impressive capabilities, but its innovation is fully realized through expert use. For example, forensic accountants and research experts interpret transactional analysis and carry out due diligence for suspicious individuals and entities. Research benefits from combining deep subject expertise with analytical approaches to derive practical intelligence from advanced technology, forensic accounting, and other fact-finding areas.
This article focuses on how advances in AI and machine learning can support forensic investigation procedures and bring detection of fraud and other financial crimes into the digital age.
The evolution of fraud and financial crimes
Fraud and financial crimes such as money laundering, digital currency schemes, and sanctions violations have challenged organizations for many years. As a result, businesses continue to focus more time and money by preventing and fighting these activities. Juniper research report on online payment fraud found that merchants and financial services organizations spend $9.3 billion a year on fraud prevention. At the same time, businesses and professional services companies are also working to improve their processes. That said, scammers continue to develop tactics to bypass new precautions. However, advances in AI and technological advances can revolutionize the detection and anti-combustion process of financial crime, making them more efficient and effective than before.
Use artificial intelligence and machine learning to detect fraud faster
AI-powered systems can analyze huge amounts of data in real time and quickly identify suspicious patterns, trends, and anomalies that could indicate malformed activity. Machine learning algorithms can continuously learn and adapt evolving fraud techniques and numerous rittanies with changing rules and regulations, enhance detection capabilities, and reduce false positives. These algorithms can make changes to processes that took quite a long time under the organizational framework, so the ability to constantly improve procedures is very powerful. Additionally, advanced technology allows for the automation of time-consuming and repetitive tasks such as data entry and document validation. This allows researchers to focus on high-value activities such as complex analysis and strategic planning.
These technological advances are strong, but the traditional organizational structure of the investigation may suppress their potential. Typically, financial crime and anti-fraud efforts are divided into two different areas: compliance and security. Isolating activities in this way can lead to inefficiency in resource allocation. Although each field has its own skill set, research efforts may track the same suspicious individuals and even use the same techniques. Companies can take advantage of potential overlap by combining efforts from both sides. This collaboration is particularly useful for applying machine learning. Machine learning models work best when given a lot of data and training across countless scenarios. This dual approach allows for the identification of fraud and at the same time allows for compliance with legal and compliance frameworks.
Data Analysis: Tech Tools or Industry Disturbance?
Robotics and automation are expected to see continuous growth due to the increasing dependence on technology around the world. In particular, over the past few years, AI has made headlines as it offers automation opportunities in many industries. Recent articles warn about the dangers of available AI technologies and focus on potential disruptions in the job market. While these concerns are unfounded (the World Economic Forum predicts that by 2030, around 30% of all jobs will be at risk of AI Automationii), the “robot acquisition” is far from the expected outcome.
AI can be used to complete intensive or repetitive tasks, but is not equipped to replace humans with tasks that require critical or expert thinking. Like other technological advances, humans have learned how to use AI to make workflows more efficient, and businesses are aware of this possibility. This shift is prominent in the information technology field when 53% of information technology experts say they have accelerated AI adoption over the past two years. III
Similarly, AI is making waves in research data analysis. This analysis often involves analysts ingesting a large amount of data, including information about financial transactions, vendors, customers and employees. The process of standardizing these data points into a unified framework is time-consuming and repetitive. The project timeline is long and drives up client costs. It is the responsibility of the analyst to maintain new and emerging tools and apply these technologies to improve client efficiency. In recent years, AI and machine learning have had several developments that allow experts to speed up and automate analysis.
Machine learning for forensic accounting and financial research
Skilled consultants use cloud-based technologies to automate the processes used to collect, validate and analyze structured (databases, transactions, etc.) and unstructured (email, chat, etc.) data sources and research algorithms. Prior to these advances in technology, data collection and verification required a significant amount of budget. Using machine learning and artificial intelligence, these times can be enhanced automated for one-time, time-intensive processes, allowing experts to focus more on data analysis and gain insights from the results.
Machine learning and AI further enable analysis and behavioral algorithms that support customized testing and flexible reporting. For example, the company suspects that employees have created false vendor invoices and diverted funds. With the right tools, investigators can bring company vendor lists, invoices, accounts to which accounts are paid, and other relevant data to an online platform to perform fraud detection analyses. They can quickly identify vendors who may share their phone number, address, or bank account number with employees and run other tests for conflicts of interest. Investigators can also set up systems that perform active monitoring along with existing internal controls. This information is available in built-in visualization tools, reducing the time required to build an analytical dashboard.
Another advance in AI provides a quick and efficient way to convert PDF data sources (bank statements, check images, etc.) into usable structured data Optimal character recognitionor OCR. This technology detects and converts text from digital files into searchable and copyable formats. Unfortunately, many OCR tools in PDF editors have been found to be unreliable, especially when text is difficult to read or when pages contain formatting inconsistencies. With an extended AI-powered OCR tool, investigators can process bank statements, checks, securities statements and other financial documents into exportable and analysable data to support their research efforts. As part of testing accuracy, these new tools compare the statement start balance with the extracted transaction details. Miscons are flagged and returned to the user to ensure accuracy.
Forensic accountants need to innovate
Throughout history, technology has shaped the field of forensic accounting, revolutionising the way financial research is conducted, and has played a pivotal role in uncovering fraud. Until recently, forensic accounting was a foreign concept for many people. Most historical research and publications focused on introducing the scope of existing fraud, and referring to the importance of organizations implementing anti-disability and detection efforts, as well as the significant cost-saving forensic accountants provided.
However, technological advances and investments by organizations provide forensic accountants with powerful tools and techniques to analyze complex financial data, detect irregularities and deliver more accurate results. Looking ahead, forensic accountants must continue to search for innovations to further change the role of research. The continuous development of advanced data analytics, machine learning algorithms, and blockchain technology enables forensic accountants to tackle new challenges in increasingly complex digital environments. These innovations enhance our ability to detect and prevent fraud, but require forensic accountants to adapt to a rapidly changing environment.
Conclusion
Fraud detection and research data analytics are two of many areas of legal and regulatory areas that are being transformed by AI and machine learning tools. Rather than rejecting these tools, employing them allows investigators to reclaim time previously spent on less repetitive tasks and instead focus on tasks that require deeper analysis and critical thinking.
AI and advanced technological tools are changing the landscape of fraud. Fraud and money laundering continue to threaten organizations around the world, and their prevalence persists as fraudsters constantly develop new methods and weaponize the same technologies that organizations can use to improve fraud prevention. However, by properly applying these emerging technologies to both fraud prevention and internal investigations, organizations can significantly improve the efficiency and effectiveness of anti-combustion responses to suit the ever-changing landscape.
1 https://www.businesswire.com/news/home/20170725005147/en/juniper-research-onley-payment-fraud-detection-spend
2 https://www.weforum.org/agenda/2020/10/dont-fear-ai-it-will-lead-to-long-job-growth/
3 https://www.ibm.com/downloads/cas/gvaga3jp
