Does Artificial Intelligence Hold the Key to Tackling Fraud?

Physical and digital threats are converging; fraudsters are evolving more elaborate and dishonest schemes to defraud companies. Rapid changes in technology are creating new vulnerabilities for fraudsters to exploit but it is often the simplest frauds that can cost companies thousands of pounds or more.

Companies are susceptible to corporate fraud at every level from the post room to the board room. Fraud is pervasive, some of the more common corporate examples include:

1. False accounting
2. Procurement/fake invoicing
3. Transaction attacks
4. Personnel management fraud
5. Identity/data theft
6. Misuse of company assets
7. Insurance fraud
8. Expenses fraud

Tackling Fraud: Where to Start?
A company will have different vulnerabilities by the very nature of its day-to-day business. Fraud is motivated by human greed. A company’s customers, employees, contractors, professional advisors, shareholders and/or non-connected third parties all pose varying threat levels. Avaricious individuals will target a company’s assets, including; cash in the bank and in hand, equipment and physical property, staff, intellectual property and customer data. A determined fraudster will find a way to exploit a company to their personal gain.

Following a risk assessment, the correct practices and procedures can be implemented to reduce the risk of fraud. It is not possible to prevent fraud entirely. Any process should include an ongoing strategy to identify new frauds and how to respond to such a threat. Staff should be trained regularly to identify and respond quickly and efficiently to any risk of fraud.

Investigating Fraud: Best Practice
All fraud leaves a trail to follow. Over the past 20 years, we’ve seen a seismic shift from a paper-based environment to the creation and storage of terabytes of increasingly more unstructured electronic data within organisations. The key to detecting, identifying, preventing further losses and recovery is the ability to analyse the data connected with the fraud quickly and efficiently.

It cannot be understated how important it is to instruct companies who are trained and experienced at completing fraud investigations at the earliest opportunity, any delay will drastically reduce the effectiveness to respond to the threat.

A traditional investigation would involve reviewing material on a document-by-document basis to identify the relevant information. With millions of electronic documents this is a long and cumbersome process.

The Role of Artificial Intelligence
Artificial intelligence (AI), in the form of digital forensic services is overcoming this issue and driving significant change in the fight against fraud. AI can do in minutes what a human would take days if not weeks to complete, at a fraction of the cost.

Relating to company data, there are three key phases in any fraud investigation: (1) recovery, (2) extraction and (3) interpretation. The likely source of the data is identified; servers, desktop computers, laptops, mobile phones and physical files. A forensic image is taken from the identified source(s). If the threat is internal this can be achieved discreetly. Suspected individuals are not ‘tipped off’ and the integrity of the data is maintained. For physical documents, these documents will be scanned, uploaded and converted into digital documents.

The data collected will be uploaded onto the digital forensic platform. A number of tools are then used to ensure that data is in a searchable form and that duplicates and irrelevant data are removed. The data is analysed to identify patterns within the data to see if there is any deletion or corruption of the data. Even attempts to delete data are indicative that somebody is trying to hide their behaviour. In certain situations this data can be recovered or a report produced detailing exactly what data has been lost, when and why.

A core set of data is then made available to search by highly trained individuals. The latest platforms enable vast amounts of data to be efficiently interrogated to investigate and evidence the fraud, including by identifying key people and applying key date and word searches. This vastly reduces the amount of data to be reviewed, producing it in a very structured and easy to review form. During the review, the AI built into the platform will learn and continue to reduce the data to remove irrelevant material, augmenting the human element in the process.

By maintaining the integrity of the data it can be used as evidence in future court proceedings against the individuals involved. The data can be analysed to prevent the same frauds repeating themselves and so the company can implement more effective fraud prevention strategies.

Artificial intelligence is a vital component to complement the fight against fraud. Effective use of technology results in a far quicker and more cost-effective solution than more traditional investigative approaches.

For further information in relation to fraud and the use of artificial intelligence, please contact Nikki Bowker or David Pack.

Share this post

Related categories



The latest news from Devonshires, sent to you direct.

Join our mailing list and find out what we’re up to and what we think about recent events and future possibilities.

Join our Mailing List