Columnists

Using AI to tackle cybercrimes

THE economic challenges that we are currently facing due to the pandemic have not only contributed towards a rise in unemployment, but could also be a factor that might lead to a rise in crime.

Although crime has dropped by 46.7 per cent during the Conditional Movement Control Order (CMCO), experts warn that the number could spike in coming weeks, as a massive increase in joblessness caused by Covid-19 will lead to more people becoming desperate and breaking the law.

It is a no-brainer that people tend to commit crimes during desperate times especially after being laid off as they lose their source of income and struggle to get by.

The emotional impact of being laid off during the Covid-19 pandemic may prove stressful and frustrating, which will make them more them inclined to commit wrongdoings.

What's more, since roadblocks will no longer be set-up and policemen will not be around as much as before to monitor the public, especially at popular hotspots during the recovery phase of the MCO, breaking the law might be seen to be easier for some.

According to a study, economic problems don't only lead to short-term adverse outcomes on the labour market, but can produce so-called "career criminals", who make ill-gotten gains through criminal activities.

In other words, the worsening opportunities in the legal employment sectors have made committing a crime more attractive for these folk. Such a tendency is expected to have more relevance to property crimes (ie. vehicle thefts) as it allows them to make some fast money.

Usually, stolen vehicles would be sold to the chop shop, which is an illegal workshop that buys stolen cars to disassemble so that their cannibalised parts can be sold for a profit.

Apart from property crime, economic theory predicts that the coming global economic recession will probably increase the amount of cybercrimes such as scams, phishing and online fraud, given the increased use of cyberspace nowadays.

If we look back to the 2008 to 2009 Great Recession (subprime mortage crisis in the US), past studies have shown a correlation between unemployment and cybercrimes, where Internet fraud increased by 33 per cent at that time. It looks like the trend will be repeated in 2020, since we are currently facing almost the same situation but with a whole new level of cybersecurity threats compared with the last one.

With the boom in unemployed graduates who are gifted in computer skills nowadays, attempting cyberattacks seems possible, what more when hundreds of tutorials and digital manuals on cybercrimes could be sourced from the Internet easily.

Recently in America, where millions of people scrambled to collect unemployment benefits and disbursements through the federal Coronavirus Aid, Relief and Economic Security (Cares) Act, the United States Security Service had reported fraudulent attempts by Nigerian criminals to claim unemployment benefits using stolen American identities.

The Secret Service warned that hundreds of millions of dollars could be lost to such scams just as states were running out of money to fund unemployment on their own. This is such a cruel act, attacking government programmes, especially during this time, where it could jeopardise the opportunity for others to receive much needed help and support.

In England, on the other hand, the Action Fraud organisation reported £5 million in losses from coronavirus-related fraud since February, with the number of reports totalling more than 2,100.

A customer from north-west England was reimbursed £102 after falling victim to a doorstep scam when a fraudster offered to do his shopping but instead withdrew cash from his card without consent.

Simon Blackburn, chairman of the Local Government Association's Safer and Stronger Communities Board, said: "People need to be cautious. If something doesn't seem right or sounds too good to be true, don't hesitate to end a phone call, bin a letter, delete an email or shut the door."

How to prevent crimes from happening?

In this digital age, Artificial Intelligence-powered crime-fighting tools could become a requirement for individuals, communities and businesses to prevent and detect crimes. The idea behind this project is that crimes are relatively predictable; we need to make use of past crimes data and use AI technology to predict future crimes.

This can be done by monitoring any behavioural changes or suspicious acts of a person so any potential crime activities can be stopped. Moreover, crime spots could also be detected beforehand using this technology. For example, a rash of thefts in one area could help predict that similar incidents might happen in the surrounding areas in the future. Thus, AI systems can aid police officers to focus on the exact location where they should consider patrolling more heavily.

In a cybersecurity context, AI technology will enable computers to recognise potential fraud cases by observing anomalies in a system with more accuracy than any human can.

The human monitor would then receive an alert for further review — whether to accept or reject this alert, before proceeding to the next course of action.

Indeed, during this challenging times, identifying suspicious activity before transactions can proceed is cheaper and more effective rather than trying to reverse them once they have taken place.

The writer is Research Assistant at EMIR Research, an independent think tank focused on strategic policy recommendations based on rigorous research


The views expressed in this article are the author's own and do not necessarily reflect those of the New Straits Times

Most Popular
Related Article
Says Stories