AI and Data Collection for Law Enforcement
By Matt Brennan
AI and data collection for law enforcement go hand in hand. It’s easy to see how the vast collection of data on individuals across the nation can be used to help prevent crime and enhance society’s greater good. Criminals use the same tools as non-criminals: smartphones, laptops, social media platforms. They leave a trail that can easily be mined for law enforcement purposes.
But there is a fine line with this sort of data collection as well. Our law enforcement AI would only be as unbiased as the data it collects. It’s imperative to make sure that the data collected is accurate and fully unbiased.
Establishing Criminal Intent
What people post on social media amounts to circumstantial evidence. It may or may not be an expression of criminal intent. It may or may not point toward resolve to commit a crime. Not all data points will be available, and therefore you’re likely not receiving a complete picture of what happens.
To understand the shortcomings of AI and data collection for law enforcement, it helps to look at another field. Researchers are developing AI to write advertising copy. The copy that is so far being produced currently falls short of the capabilities of a human copywriter that understands how to play on human emotion. The same thing holds true in law enforcement.
Criminal investigations and law enforcement involve more than a simple collection of data points. Eyewitness accounts, 911 recordings, and police body cam footage factor into criminal investigations. The human element is critical here for AI to develop an accurate track record for predictive policing.
While AI and data collection will become increasingly important tools for law enforcement, a human connection will always be vital in this field.
Data Breaches From Police Departments
Law enforcement is not immune from data breaches. As predictive policing and investigation AI are more heavily implemented, police departments will find themselves housing data in larger volumes. This makes them a bigger target for data breaches.
In June of 2020, hundreds of thousands of potentially sensitive files from police departments across the US were leaked online. The leak included 10 years of data from over 200 departments, fusion centers, and other law enforcement training and support resources.
The Los Angeles Police Department was also recently hacked, exposing the confidential personal information of up to 20,000 individuals. Names, dates of birth, email addresses, and passwords were all made public.
Data Collection for Law Enforcement
As police departments collect massive volumes of data to use in their investigations, they will need to safeguard that data and protect the people involved. Data collection for law enforcement will become more prominent in the near future. AI and machine learning will depend on this data in order to enhance the investigatory process. But safeguards need to be taken in order to ensure safety – especially when people may not have willingly given their personal data over. Police departments can back this data up, and work to encrypt it. They can make sure that it is password protected, and only accessed from secure networks.
Any of these steps may work to improve the security of data.