Generative AI, which is
The use of artificial intelligence in anti-fraud programs is expected to triple over the next two years, according to a report released Tuesday by the Association of Certified Fraud Examiners, and the use of generative AI in particular is projected to rise 83% in the same timeframe.
These technology investments will bolster the 18% of professionals who currently count artificial intelligence and machine learning among their fraud-fighting tools. Another 32% anticipate implementing these technologies in the next two years. Meanwhile, more than 50% of current anti-fraud programs already use or are expected to adopt related technologies like computer vision analysis, robotics and behavioral biometrics at some point in the future. These are all part of the wider use of data analytics technologies that have become a major point of interest in the anti-fraud community.
However, the ACFE noted that many organizations struggle with implementation, despite their stated intentions.
"Despite the expected increase in the use of every data analysis technique in our study, reported adoption rates have shown little growth since 2019, highlighting the slow pace at which organizations are able to implement new technologies," said the report.
The most common reason for implementation difficulties is budget. Finding the financial resources to implement anti-fraud technology was cited as either a moderate or major challenge by 82% of the professionals polled. Following this were staffing/in-house skill limitations (74%), concern over poor data quality or integration (73%), lack of perceived ROI (68%) and data governance/transparency concerns (60%).
Despite, or perhaps because of, these budget concerns, 59% of the respondents said they plan to increase their anti-fraud technology budget in the next two years. In contrast, only 6% said they plan to decrease budgets.
Guarding against payment fraud highest priority
The survey also contained data about specific types of fraud. When asked which risk areas are being monitored for fraud, 44% of the respondents cited disbursements and outgoing payments fraud, 41% said procurement or purchasing fraud, 32% said fraud by customers/first party fraud, 31% said travel and expense payment fraud, and 30% said fraud involving thefts or incoming payments.
The lowest priorities were inventory theft (21%), hacking or data breaches (same), identity theft/account takeovers (22%), corruption/bribery (25%), and fraud committed by vendors or contractors (26%).
As for where they get the data they analyze, the vast majority (77%) turn to internal structured data. The next closest source, at 40%, is public records, followed by internal unstructured data at 33%, law enforcement or government watch lists at 31%, social media at 28%, other third party data at 25%, and data from connected services, also 25%.
The report is based on a sample of 1,187 anonymous ACFE members, who were polled in October 2023.
