Understanding event chronology, specifically on information exchange using mediums such as, but not limited to, emails, SMS and other messaging platforms, call records and social media communicationsBy Nikhil Bedi & Vivek BhamodkarWith evolving business models, increased use of tech and a changing regulatory landscape, fraud management is fraught with newer and more complex challenges. These are further exacerbated during cross-border investigations, where varied levels of standardisation, languages, local laws and regulations, along with specific cultural attributes, bring additional complexities—mandating an investigation methodology standardisation and requiring tools for quick insights.
While cross-border investigations could relate to bribery and corruption, FCPA, embezzlement, conflict of interest, data breach, IP theft, financial reporting fraud, etc, these investigations often entail the following procedures:—Drawing inferences by correlating external intelligence (open source information, adverse media, subscription services) and internal intelligence (business operations, transactional data, employees and vendors);
Related News
—Extracting information from varied documents and formats in a timely manner;
—Establishing a relationship between key entities and individuals;
—Understanding event chronology, specifically on information exchange using mediums such as, but not limited to, emails, SMS and other messaging platforms, call records and social media communications; and
—Understanding processes and generating insights from transactional data.
Emerging technology and artificial intelligence (AI) can help make investigations efficient, generate insights, and/or aid reviews. Optimal use of AI demands the knowledge of ‘possibilities and limitations’ of such techniques, either in the form of ‘special purpose software’ or the ability to combine various methods. While there are numerous areas for AI application, some solutions/techniques worth mentioning are:
—Identification of relevant content: Embedded AI in specific e-discovery and data analytics platforms helps narrow down content by generating interactive document visualisations and classifying them. Semantic search automatically identifies related concepts and documents by training the tool using select documents and reducing noise;
—Digitisation, data extraction and automation: Advanced computer vision algorithms offer the ability to extract information from bank statements, invoices, waybills and various other documents, and are often combined with natural language processing (NLP) techniques to extract entities from documents and perform automated verification using digital channels. Verification of tax details, KYC documents, incorporation details, etc, may be expedited using a combination of AI techniques;
—Advanced analytics: These have extensive applications in investigations, including:
1. Sentiment analysis: It can differentiate documents based on tone, subject, positivity or negativity, and helps identify suspicious conversations, reviews, news articles, etc.
2. Entity extraction: AI can help discover additional undisclosed entities referenced in documents, emails, bank statements, etc, by applying entity extraction methods.
3. Network analysis: It’s a powerful medium for analysing information flow and connecting the dots in an investigation to help establish relationships between individuals and entities based on business dealings, reference information and reporting hierarchy.
4. Outlier scanning: Supervised and unsupervised learning algorithms can help detect anomalies and outliers from transactional data.
—Process analytics: Some recent applications of AI allow users to visualise a ‘metro-map’ of business processes, which can help link undesired process behaviour, and further scrutinise underlying users and transactions for non-compliance or violations.
As significant a role as these tools and technologies may play in the fight against fraud, use of AI, NLP and other technologies come with their own set of challenges. One must consider scalability, the ability to reproduce results consistently, and prior knowledge of these tools for their efficient use. Further, emotion and instinct remain exclusive to mankind and AI has not been successful in replicating this so far.
While cross-border collaborations and global settlements increase in frequency, technologies that reduce manpower and improve efficiencies become more relevant. With over 5% estimated revenues being lost to fraud each year (2020 Report to the Nations: Asia-Pacific Edition), and international crimes gaining complexity and chronicity, it’s time we equipped our investigators with the right AI tools and technologies to tackle these situations.
Bedi is partner & leader, and Bhamodkar is director, Forensic, Financial Advisory, Deloitte India
Get live Stock Prices from BSE, NSE, US Market and latest NAV, portfolio of Mutual Funds, Check out latest IPO News, Best Performing IPOs, calculate your tax by Income Tax Calculator, know market’s Top Gainers, Top Losers & Best Equity Funds. Like us on Facebook and follow us on Twitter.
Financial Express is now on Telegram. Click here to join our channel and stay updated with the latest Biz news and updates.
Source: https://www.financialexpress.com/opinion/using-artificial-intelligence-for-forensic-probe/2222690/
