Artificial intelligence (AI) is being used both to monitor and prevent crimes in many countries. In fact, AI’s involvement in crime management dates back to the early 2000s. AI is used in such areas as bomb detection and deactivation, surveillance, prediction, social media scanning and interviewing suspects. However, for all the hype and hoopla around AI, there is scope for growth of its role in crime management.

Currently, a few issues are proving problematic. AI is not uniformly engaged across countries in crime management. There is fierce debate on the ethical boundaries of AI, compelling law enforcement authorities to tread carefully. Defining the scope and boundaries of AI, which includes personal data collection, is a complex task. Problems notwithstanding, AI represents a promise of a new paradigm in crime management, and that is a strong case for pursuance. (For more on crime-fighting tech, see 4 Major Criminals Caught by Computer Technology.)

What Is the Crime Prevention Model?

The crime prevention model is about analyzing large volumes of various types of data from many different sources and deriving insights. Based on the insights, predictions can be made on various criminal activities. For example, social media provides a veritable data goldmine for analysis – though, due to privacy concerns, it is a contentious issue. It is a known fact that radicalization activities by various groups are done through social media. AI can reveal crucial insights by analyzing such data and can provide leads to law enforcement agencies.

There are also other data sources such as e-commerce websites. Amazon and eBay can provide valuable data on the browsing and purchasing habits of suspects. This model is not new, though. Back in 2002, John Poindexter, a retired admiral of the U.S. Army, had developed a program called the Total Awareness Program which prescribed collecting data from online and offline sources. But following vehement opposition due to privacy intrusion issues, funding support to the program was stopped within a year. (To learn about fighting cybercrime, check out How I Got Here: 12 Questions With Cybercrime-Fighter Gary Warner.)

Real-Life Applications

AI is starting to be used for crime prevention in innovative ways around the globe.

Bomb Detection and Deactivation

The results of deploying robots in detecting bombs have been encouraging, which has led to the military procuring robots worth $55.2 million. Over time, robots have become more sophisticated and can distinguish between a real bomb and a hoax by examining the device. According to experts, robots should soon be able to deactivate bombs.

Surveillance, Prevention and Control

In India, AI-powered drones are used to control crowds by deploying pepper spray and paintballs or by making announcements. Drones are fitted with cameras and microphones. Drones, it is believed, will soon be able to identify people with criminal records with facial recognition software and predict crimes with machine learning software.

Social Media Surveillance

Social media provides the platform for executing different crimes such as drug promotion and selling, illegal prostitution and youth radicalization for terrorist activities. For example, criminals use hashtags to promote different causes to intended audiences. Law enforcement agencies in the U.S. have succeeded to an extent in tracking such crimes with the help of AI.

Instagram, for example, is used to promote drug trafficking. In 2016, New York law enforcement used AI to track down drug peddlers. AI searched for millions of direct and indirect hashtags meant to promote drugs and passed on the information to police. Similarly, to tackle radicalization of youth, law enforcement agencies are using AI to monitor conversations in social platforms.

Interviewing Suspects

An AI-powered chatbot in a university in Enschede, Netherlands is being trained to interview suspects and extract information. Expectations from the bot are to examine the suspect, ask questions and detect from the answering patterns and psychological cues whether the suspect is being truthful. The name of the bot is Brad. It is still in the beginning stages, but the development represents a new aspect in crime management.

Advantages and Disadvantages

While these futuristic advances in law enforcement have a lot of potential, one must also consider the drawbacks.

Advantages

Security needs and considerations are dynamic and complex, and you need a system that adapts quickly and efficiently. Human resources are capable, but have constraints. In this view, AI systems have the advantage of being able to scale up to do their jobs more efficiently. For example, monitoring possible criminal activities on social media, from a manual perspective, is a gargantuan task. Human approaches can be erroneous and slow. AI systems can perform this task by scaling up and performing the tasks faster.

Disadvantages

Firstly, for all the hype around, AI’s involvement in crime management is still in the nascent stage. So, cut the hype and accept that its efficiency in crime prevention or control on a larger scale is still unproven.

Second, crime prediction and prevention will require data collection, much of which could be personal data. This makes the government and law enforcement agencies vulnerable to extreme criticism from citizens and other groups. This will be interpreted as intrusion on citizens’ freedom. Data collection and snooping have been extremely contentious issues in the past, especially in democratic countries.

Third, developing AI systems that learn from unstructured data can be an extremely challenging task. Since the nature of criminal activities have been becoming more sophisticated, it might not always be helpful to provide structured data. It is going to take time for such systems to adapt.

Conclusion

Currently, there are many challenges confronting the involvement of AI systems in crime management. However, it is worth the effort to engage AI in crime prevention and control. The nature of crime and terrorist activities is evolving to become more sophisticated every day, and purely human involvement is no longer enough to tackle such problems. In this context, it may be important to note that AI will not replace human beings, but will complement them. AI systems can be fast, accurate and relentless – and it is these qualities that law enforcement agencies will want to exploit. As of right now, it seems that AI will continue to become even more prominent in law enforcement and crime prevention.