Across the country, law enforcement agencies are using emerging tech to gather an unprecedented amount of data to drive down crime statistics. But are their efforts actually making a difference?
In 2017, the Chicago Police Department created six high-tech police hubs located throughout the city’s more crime-ridden neighborhoods. Dubbed “Strategic Decision Support Centers,” the hubs are a blend of human expertise and high-end technology, including surveillance cameras, gunshot detection platforms, predictive mapping and data analytics.
Since the centers went live, crime in adjacent neighborhoods has gone down. In two of the districts that have had some of the city’s highest crime rates, the decline in crime rates has been so significant, the numbers are helping to drive down the city’s overall crime numbers. “Before this project started, I would have said there’s no way technology can have this kind of impact,” said Jonathan Lewin, chief of CPD’s Bureau of Technical Services. “But it does.”
Chicago’s experience highlights the advances that law enforcement has made not just in adopting new, high-tech crime-fighting tools, but also with integrating the various tools, systems and platforms and then turning massive amounts of data into intelligence that can make cities safer. Some of Chicago’s technology has been around for a while. Some of it is new. What’s different is how CPD has optimized and integrated it with human experts, according to Lewin. “There’s a daily intelligence cycle we didn’t have before that leverages all of this information,” he said.
But, as experts will tell you, it’s too simplistic to suggest that technology is solving the country’s crime problem. “Technology is a piece of a larger puzzle, and so you can have the greatest piece of technology, but if you don’t implement it well, it’s not going to have its intended effect,” said Dave McClure, a research associate at the Urban Institute. “The good news is that the police are getting better at turning data into useful information.”
For anyone who remembers the surge in crime, especially murders, during the 1970s and ’80s, the precipitous drop in the crime rate over the past 25 years is nothing short of remarkable. The national crime rate peaked in 1991 at 5,856 crimes per 100,000 people, according to the Brennan Center for Justice. By 2016, the overall crime rate had declined to 2,857 offenses per 100,000, less than half of what it was in 1991. While some cities have seen increases in violent crime and murder in recent years, there is no evidence that the public safety gains of the past 25 years are being reversed, according to the center.
Records Management: Old Tech Gets a Makeover
They are neither trendy nor cutting edge. But they could be called the workhorse of law enforcement. Records management systems have been in the back rooms of law enforcement agencies for many years, but Dave McClure, research associate for the Urban Institute, argues they can have a major impact on police operations that will grow as better standards make the information they contain more useful in reducing crime.
“The original purpose of records management systems was to pull up notes about individual cases,” said McClure. “But interest grew to make comparisons across cases.” Unfortunately, many of the original systems weren’t designed to do that. However, police agencies now see the value in sharing information between agencies, because crime doesn’t stop at a city’s border. “Data sharing between agencies’ RMSs will turn police data into a commodity,” he said. “Instead of being agency specific, police data can now be mapped across systems and shared. That will open up unbelievable possibilities.”
Progress around data standards is advancing, according to McClure, but more needs to be done. Data sharing between agencies will be a heavy lift, however. That’s because individual cases have unique characteristics that may be hard to standardize. But the payoff could be huge: lower RMS costs overall, because there’s less likelihood of vendor lock-in for agencies, and less likelihood of criminals and their crimes slipping through the cracks.
During this period of declining crime rates, city budgets for policing have steadily grown. In 1977, state and local governments spent $58 billion on police and corrections, according to the Urban Institute. Today, the U.S. spends $100 billion on policing and another $80 billion on corrections, according to a 2017 report from Statista, an online statistics and market research firm.
What individual cities spend on policing varies significantly. Oakland, Calif., spends 41 percent of its general fund on policing, while New York City spends a modest 8.2 percent on law enforcement, according to Statista. Whatever the policing expenditure, city and county law enforcement agencies have increased their investments in technology. Globally, law enforcement agencies are expected to spend $11.6 billion on software tools and systems, according to MarketsandMarkets, an online B2B market research firm. Spending is expected to grow at an annual rate of 9.3 percent, reaching $18.1 billion by 2023.
While figures on what U.S. local and state governments spend on policing technology are hard to come by, the global growth in law enforcement tech spending has been driven by evolving policies that focus on community policing and by advances in software for mapping, various types of surveillance and analytics. For cities that can afford the new policing technology, it’s the dawning of a new era.
When Maggie Goodrich was CIO for Public Safety with the city of Los Angeles, the joke was that the Los Angeles Police Department was a big data company disguised as a law enforcement agency. “We are inundated with data,” she said. Goodrich, who is now chair of the Public Safety Technology Alliance, pointed out that the key to making sense of the burgeoning data is distilling it down and deciding what’s important in the moment. Doing that won’t be easy, however. Policing data continues to pile up, thanks to technology that’s better and faster than ever.
Take a look at the new technologies and it’s easy to understand why police departments are flooded with both structured and unstructured data. Brian Jackson, senior physical scientist with RAND Corp., called the growth in data, databases and information sharing one of the most significant trends in policing technology. “Look at the growing number of alert systems, video feeds, especially body cameras, to understand the proliferation of police data and the need to manage that flood of data,” he said. There are other reasons why police data is exploding. “One arrest of a person whose computer is a piece of evidence might result in terabytes of data for just that one case,” Jackson explained.
If there was just one piece of technology that represents today’s new, data-driven police force, it would be the body camera. The miniaturized, mobile video tech tool has taken off ever since the riots in Ferguson, Mo., in August 2014 led President Obama to push through $263 million in funding for police body cams as a way to bring transparency to police interactions with people. Today, most major police departments use the devices to record these interactions. The market is dominated by Axon, formerly Taser International, and now the largest producer of police cameras, especially since it acquired rival Vievu in 2018. Body cams also represent a technology that has changed the relationships between officers and the public, with a growing body of evidence showing respect by officers towards those with whom they interacted — a key purpose of the technology.
Gunshot detection systems represent a different technology trend that has benefited the police. By combining sensors — an array of microphones — with spatial mapping, police have a new way of responding rapidly to violent incidents. With the rise of gun violence and an increase in illegal guns, city police have often been one step behind when a spate of gunshots rings out in a neighborhood. Gunshot detection technology, the most notable offering from ShotSpotter, offers a faster and more accurate response than to 911 calls, say experts. As of September 2018, 95 cities in the U.S. and South Africa were using ShotSpotter’s technology, according to the company.
Sensors that can pinpoint gunfire are just one kind of surveillance the police can now use. Video surveillance has been around for a while, but advances in technology have magnified its capabilities. Chicago has built the largest municipal camera integration platform in the country, according to CPD tech chief Lewin, with more than 35,000 government and private-sector video cameras on tap to watch and record what is happening on the streets of the Windy City. Another technology — license plate readers — uses character recognition to read the numbers and letters on license plates and quickly compare the plate information with hotlists of stolen cars, or drivers whose licenses have been suspended or revoked.
While LPR technology is relatively mature, police have continued to evolve how they use it in a way that is effective yet doesn’t interfere with privacy concerns. In Denver, for example, police can only view a live feed of LPR data — no past records are kept. City police are also careful not to target specific neighborhoods, but to deploy the technology throughout the city on cruisers or in stationary locations.
Less mature, but certainly more transformative, is the growing field of crime analytics, including predictive policing and artificial intelligence. As the amount of data available to law enforcement increases, the need to turn it into information and ultimately intelligence has opened the door to analytics tools, including some AI techniques, that can automate certain human tasks.
“AI is the next logical evolution in policing,” said CPD’s Jonathan Lewin. “We have all this data, a lot of sensors, and incoming information from other open sources, including crime tips from citizens. So, plugging all of this into some kind of engine to gain insights and make connections that wouldn’t be obvious to a human is the next logical step.”
Jonathan Lewin (right), chief, Bureau of Technical Services, Chicago Police Department
Chicago is working with technology from Microsoft and Genetec, a Canadian firm, and has built a high-end, integrated decision support system that is giving CPD’s support centers the insight needed to reduce crime. New York City has constructed a similar platform, known as the Domain Awareness System, which turns big data from sensors, cameras, license plate readers and other devices into actionable information for the cops.
Less comprehensive, but just as leading edge, is the use of chatbots to automate some of the work done by police dispatchers. The San Diego County Sheriff’s Department has launched a bot that helps deputies receive information while in their cruiser. Normally, deputies call their dispatchers to check on license plate numbers or run a profile check on a suspect. The department has been working with Microsoft to allow deputies to access the same information via a voice-activated assistant, which can pull the information from back-end databases and “tell” the officers what they need to know in real time.
While these trends represent some astonishing leaps in the use of technology to fight crime, it might sound somewhat hyperbolic to say an even bigger trend is about to sweep through public safety. Yet FirstNet, the high-speed, wireless, interoperable network for first responders, could be just that. First envisioned in the wake of the terrorist attacks of 2001, FirstNet went live in 2018, and promises to deliver big data at high speeds while providing the police with priority and data pre-emption along with heightened security.
“It is hard to underestimate the importance of having such a powerful network that can allow so much more data to be exchanged so quickly,” said McClure. “FirstNet is going to really dramatically change the way the police can act as information collectors and transmitters.”
FirstNet could be the tipping point as far as policing and mobile technology is concerned. From devices used by officers in the field to wireless cameras anywhere and everywhere, including on drones, the police will no longer be tethered to physical stations. “We are talking about the officer becoming the hub for everything that goes on, rather than it being a station or vehicle,” said Goodrich.
Public safety communications has been hit by network throttling, a problem that FirstNet is designed to overcome. The increased use of mobile technologies and the growing reliance on sending and receiving data in real time during emergency situations makes the buildout of FirstNet and its 5G networking capabilities important for both tactical and strategic purposes.
FirstNet has been funded by Congress with nearly $7 billion in seed money, while AT&T, the network’s provider, has said it would spend roughly $40 billion during the 25-year contract with FirstNet to construct, run and maintain the network. It’s a clear example of backing a new public safety technology with adequate money. But that’s not the case for the nation’s nearly 18,000 local law enforcement agencies where funding and the high cost of new technology remains a stubborn hurdle.
“Cost is always an issue,” said Lewin. “Everybody has limited budgets. ShotSpotter has a real impact, but it’s also not cheap. Surveillance cameras have a huge impact, but they aren’t cheap either. It’s one thing to make the initial investment, but sustaining the investment is critically important.”
Law enforcement agencies often rely on grants to set up programs that take advantage of new technology. But once that grant money runs out, the program, along with the tech tools it uses, can be difficult to maintain financially. ShotSpotter charges between $65,000 and $90,000 per square mile, per year, according to Forbes.com. Licenses to use LPR technology can cost tens of thousands of dollars annually. Another ballooning cost is data storage, especially for body cameras. Depending on the city’s retention policies, those costs, which were zero a few years ago, are now a major factor.
While tech firms will point to the cloud as an affordable storage option, RAND’s Brian Jackson pointed out that cloud-based storage raises questions and concerns over the control of the data. “Another pressure point is ‘tech lock-in,’” he said. “When a law enforcement agency stores data in the cloud, do they have the ability to change who stores the data and where they store the data?”
The Public Safety Technology Alliance, which Maggie Goodrich chairs, has made the establishment of open standards along with interoperability its core mission. For years, law enforcement has struggled with a lack of open standards that has led to a growing number of technology islands. But FirstNet, which was established in part because of the lack of interoperability for first responder communications during the World Trade Center attacks, has begun to shift the conversation, according to Goodrich.
“If vendors are going to deliver data across FirstNet and to police departments, it’s got to be done with open standards, so that law enforcement can exchange information, whether it’s in the form of voice, data or video,” she said.
T.J. Kennedy, CEO of the Alliance and former president of FirstNet, echoed the remarks by Goodrich and pointed out the advantages for law enforcement down the road. “If we have open standards in law enforcement, it’s going to drive the economies of scale and it’s going to drive innovation by industry, which knows what the open standards are that we need to follow,” he said. “That innovation and economies of scale means we will have better technology that’s going to be cost-competitive going forward. That is something we’ve not had in the past.”
No matter which policing tech trend seems the most innovative or cost-effective, the ultimate measure of its success will be in outcomes that impact crime while not impacting privacy or civil liberties. “The challenge, as far as using technology to drive down crime, is always in the application,” said Jackson. “If not done right, the technology will have less of an effect. It always comes down to implementation.”
CPD’s Lewin likes the outcomes he has seen when various technologies, tools and platforms are integrated together and deliver intelligence. “What’s tough to say is what part of the technology process is having the biggest impact,” he said. “But some combination of everything seems to be the answer.”