Internal investigations can decide the fate of companies and individuals. A few of the interesting ones are listed below:
- Deutsche Bank AG’s internal investigation, Project Teal, found that certain employees in Spain were guilty of misselling currency derivatives. A number of management executives in Spain and London were replaced as a result of the internal investigation.
- In a personal setback, Mark Hurd of HP was let go after an internal investigation because of sexual harassment charges filed against him in 2010.
Why Are Internal Investigations Conducted?
Internal investigations could be triggered due to various factors, including suspected breaches of regulatory duties or internal compliance rules.
Regulatory or criminal investigations by government authorities, whistleblower tips, employee misconduct, findings from audits, and information from employees or external parties can also ensue internal investigation.
The process of investigation starts with identifying the matter triggering the investigation. Next, the investigators define the scope of the investigation, its objectives, and methods, including identifying key witnesses and gathering evidence.
Once the collected data is analyzed, the findings are shared with required parties, for example, the senior management, legal counsel, and relevant stakeholders.
As in life, internal investigations rarely follow the path that it is supposed to on paper.
Challenges in Conducting Internal Investigation
Although the process sounds straightforward, conducting an internal investigation is complex.
Investigations are often stalled if the board or management fears a chance of uncovering illegal activities, potentially triggering legal actions against the organization or individuals involved. Employees and other parties might be reluctant to cooperate during investigations for fear of retaliation.
Resource limitations, conflicts of interest, and organizational cultures that tend to tolerate misconduct can also hinder investigations.
High-profile investigations can attract media attention. The pressure to respond quickly to allegations in the public eye can compromise the thoroughness and accuracy of internal investigations.
Additionally, navigating legal privilege complexities and addressing the potential erosion of employee trust and morale are challenges that need to be dealt with kid gloves during an internal investigation.
Investigators need technical expertise to analyze digital evidence effectively in cybercrimes or digital misconduct cases.
Preserving the integrity of evidence, especially digital data susceptible to alteration or deletion, presents another significant challenge.
Concerns about data privacy may be raised, as investigators may need to access sensitive personal information that could breach data protection regulations.
Although various organizational and technical obstacles remain, internal investigations have evolved over the years to encompass the changing face of businesses.
Evolution of Internal Investigations with Changing Business Environments
Traditionally, internal investigations were conducted manually, relying on paper records, interviews, and limited data analysis capabilities. These investigations were often time-consuming, costly, and prone to human error.
Internal investigations have evolved in response to changing business environments and technological advancements. Global businesses meant dealing with multiple jurisdictions, languages, and diverse cultures, which also brought growing complexity of business regulations and the need to stay compliant. With a significant part of businesses being digitized, investigating meant efficient analysis of large volumes of data generated due to these digital processes.
In navigating changing business environments and the challenges faced during an internal investigation, technical advances in AI offer significant advantages. Some areas where AI is already helping are:
Early Case Assessment
AI tools are helping investigators run an automated Early Case Assessment (ECA) to determine the most relevant information at the outset of an investigation matter. This helps determine potential legal liability, calculate risks, and estimate project costs.
Structured and Unstructured Data Analysis
Reviewing large volumes of unstructured data, like emails and scanned hard copies, using AI to find and prioritize information with the help of keywords and patterns can streamline internal investigations.
Continuous active learning (CAL) helps the AI engines adapt and rank data based on relevance, while concept clustering is used to identify contextual similarities and create logical groupings from available sources.
Additionally, data points from the analysis of unstructured data can uncover patterns in structured data. For example, during internal investigations, if a suspected customer is identified in an email, AI can identify similar customer accounts based on shared features, like addresses and onboarding timelines.
Automating Repetitive Tasks
AI efficiently spots confidential details buried in unstructured content, like scanned documents, and redacts personally identifiable information (PII) and sensitive personal information (SPI). Thus eliminating the need to import spreadsheets for tedious pattern matching. This enhances security and saves valuable time in the redaction process.
During internal investigations, identifying whether multiple records refer to the same real-world entity, such as a person, organization, address, or phone number, is crucial. This process, called entity resolution, uses a combination of NLP and AI. NLP helps make sense of unstructured data to connect entities even if they are in different languages, while AI sifts through identification numbers, addresses, and email IDs to find links. Entity resolution using AI and NLP is more accurate than manual processes or fuzzy logic systems.
Digital forensics assists in identifying and collecting digital evidence, analyzing it, validating its integrity, and compiling a forensic report for internal investigation purposes. In the case of corporate fraud like Theranos, perpetrators not only create complex schemes to deceive organizations but also use data manipulation techniques to hide illicit activities. Fraudsters often create false documents like financial records, receipts, and invoices to create an illusion of legitimacy or manipulate them.
A specialized digital forensic solution powered by deep learning can detect fake documents by analyzing key information extracted from documents, metadata inconsistencies, and pixel manipulation of images.
No wonder traditional, manual methods used during internal investigations are being augmented with AI-powered tools that facilitate early case assessments, efficient data analysis, task automation, and digital forensics.
It’s time to reinvent your internal investigation practice and adopt the latest technological innovations. Contact Knovos to learn how our in-house built technology solutions to modernize your investigation practice.