In two years artificial intelligence and machine learning will be used for the anti fraud three times more often than today. Such data were obtained in the course of a joint study of SAS company and the Association of Certified Fraud Examiners (ACFE). At the moment such anti-fraud tools are used in 13% of companies, participated in the survey, and other 25% said that they plan to implement them during next 1 or 2 years.
The joint study of SAS and ACFE was launched in February 2019 and its results were summarized in the end of June, 2019. Answers of 1055 members of ACFE, working in different countries, were investigated. Following the results the interactive report was created. Questions to experts were about technologies and tools, which are used in their organizations to fight the fraudulent activity.
As it turned out, today the majority of companies often use pre-configured reports on key fraud events with the use of classical tools, for example, from Microsoft Office. This is a basic tool for 64% of respondent companies. The 2nd place goes to the automatic monitoring with the use of expert business rules – 54% of companies use it. The visual data study with the use of BI tools closes the three – 35%.
In addition to the growing interest in AI, the authors of the study identified the following trends: widespread distribution of biometrics. Nowadays it is used by the each 4th company and 16% of respondents want to implement it till 2021; more attention will be paid to different tools and methods of data analyzing. By 2021, 72% of companies will use the automatic data monitoring, automatic anomaly detection and so on to combat the fraud; predicative analytics is going to be used in 42% of organizations, what is 22% more than results of the previous study; the growing trend in the use of data visualization tools continues – they are used or will be used by 47% of companies (today nearly 35%).
One of the questions, asked by participants, concerned the most frequently used analytical tools. It turned out that SAS solutions are usually chosen for tasks of predicative analytics and modeling, for analyzing social relationships and use of graph analytic, for deep analyzing of text information.
"As attackers use the latest technology to build more perfect criminal schemes, professionals who oppose them must also resort to the most advanced solutions, – said Bruce Dorris, the President and CEO of ACFE. – The answer to the question of who will be the winner of this confrontation depends on the choice of the most effective technologies and risk management tools".