Page 37 - Understandinging Forensic Technology Landscape

RPA requires human monitoring to address breaks is, home fixtures, appliances, automobiles, phones,
or exceptions that come from the automation and and wearable devices) through sensors, computing
maintenance over time. Often, RPA is employed when applications, and connectivity technology (often
system-based automation of tasks is not feasible or Bluetooth or Wi-Fi) to collect, process, and share data.
needs to be proven out with a test case. • Augmented Reality (AR) and Virtual Reality (VR) — AR

Artificial intelligence (AI) contextualizes digital information to provide a visual or
AI is a broad category of technologies that can perform audio overlay across the physical world. VR provides
tasks requiring elements of decision-making, perception a computer-generated audiovisual, three-dimensional
and translation in software programs. AI includes the reality. Current examples are AR “smart glasses” and
following: experiential training/simulations via VR.
• Machine learning (ML) — A type of AI in which • Blockchain — This emerging technology is discussed
40
software algorithms learn from their experience separately in this reference guide.
and become more adept at performing a task with
additional iterations. ML is further sub-categorized as Common use cases for RPA, AI,
“supervised,” where the model’s outcomes in test or
seed pools are reviewed by humans to train the model and emerging technologies
for future data points and “unsupervised”, where Forensic applications
outcomes are unknown or available and systems Businesses are using RPA to automate financial
identify clusters, associations, or anomalies to develop processes inclusive of data entry, data migration and
a model. ML models are often tailored to specific data validation tasks. Applications exist throughout
problems or pools of data. businesses but often appear first in routine financial
• Language Processing — AI applications that interpret reporting functions such as the following:
unstructured (text-based) data. Rather than searching • Source-to-cash processes
for exact terms, language processing finds near
matches or identifies content that fits previously • Procure-to-pay processes
defined parameters in similar or disparate forms. • Data reporting and dashboarding
Language translations, speech-to-text capabilities, and • Reconciliations across financial systems
text-to-speech capabilities are examples.

• Conversational Interfaces — AI that applies the Automation enhances the work of forensic practitioners
patterns and conventions of human conversation by making the processing and analysis of data more
through text or audio. These tools interpret inputs and efficient. For example, RPA is being used in forensic
determine appropriate responses without following a engagements to reformat massive amounts of data
set of specific, pre-defined business rules. collected in the discovery process so that the data can
be loaded into databases and analyzed. RPA saves
Emerging technologies a tremendous amount of time and improves quality
It is important to consider that automation, with all of its compared to having an analyst manually copy and
unique intricacies, is only one area of technology that is paste data from one spreadsheet into another and
transforming how we do business. Additional emerging then reformatting the data. Practitioners benefit from
technologies include the following: a working knowledge of existent automation programs
• Internet of Things (IoT) — These technologies relate and basic scripting so that they can automate their
to the interconnectivity of physical products (that own tasks.

40 For more information, see Blockchain section of this Guide

Understanding the forensic technology landscape | 33