Getting Ahead of Trouble

Artificial intelligence for safer schools

• By John Carter
• June 01, 2019

Everyday nearly 75 million students attend school and rely on administrators to keep them safe. Unfortunately, too many tragic school shootings have shown educators and parents that what is currently in place is simply too little, too late. Much more needs to be done to protect our children. The good news is that in 2019 alone, lawmakers across the country introduced nearly 250 school safety bills to help prevent or mitigate the next mass tragedy on a campus.

Recently in Texas, lawmakers held hearings on safety proposals among schools and students impacted by gun violence. The proposals that earned the most support of lawmakers included those that strengthen security and mental health initiatives in schools. They also agreed that campuses need to have programs in place to ensure a quicker response when school shootings happen, and that lawmakers need to take steps to help prevent those shootings from happening in the first place.

On the national level, the Trump administration’s 2020 budget includes a $200 million increase in funding for school safety with programs aimed to prevent, mitigate or respond to violence. Thankfully, school safety issues are front and center at both the state and national levels. While tremendous progress is being made, more action can be taken at the local school district level to ensure school safety by modernizing physical security with the help of artificial intelligence.

Connecting the Dots with A.I.

According to Benjamin Franklin, “an ounce of prevention is worth a pound of cure.” The programs and bills all seek to find a cure to school violence, but the key is in early detection and warning signs. Artificial intelligence, and its ability to integrate predictive insights with physical security, will be a game-changer for protecting the lives of students, by preventing threats before they happen.

Artificial intelligence (A.I.) is based on an Artificial Neural Network and works like the human brain. Just as humans have senses to see, hear, smell, taste and feel—so too does an artificial neural network. Like the human brain, it takes in a wide variety of inputs and data over time and constantly evaluates it to discover, predict, and make decisions based on machine learning. The more connections and integrations it has with security systems and non-security applications and data, the more informed and accurate it becomes.

Already, forward-thinking schools are experimenting with various forms of A.I., beginning by monitoring students and behaviors through new technologies. For instance, school safety apps monitor student attendance on buses, take in information about student behavior from bus drivers and track important markers such as student fights, bullying, or misbehavior. Data on students is collected, reported if needed, and goes into a larger system.

Teachers can monitor and record daily behaviors, such as did the student act up, were the parents notified, is there a problem at home, did he see the counselor or principal? Did the parents respond? Collecting data overtime can create a risk score for students, and determine if threatening behavior is increasing or suspicious, if a student needs attention and can potentially alert school officials before a bigger problem arises.

Schools are also employing other A.I. applications that scan and monitor emails, texts and social media to look for and flag signs of violence, depression, drug and alcohol use. “See something say something” apps, allow users to report tips on suspicious activity to their school district, school system and law enforcement all at once.

There are usually telltale signs of trouble before a mass shooting occurs, but it must be detected to be acted upon. While the apps are a tremendous improvement and in many cases spot threats before they arise, data in isolation is not effective. School safety administrators must make it a priority to tie all the data together and connect all the dots in order to make the most intelligent safety decisions and have the ability to prevent or mitigate in active school shooter situations as early as possible.

Acting Quickly to Protect What Matters Most with Risk-Adaptive Access Control

Tools now exist through A.I. assistance and integration allowing schools to automatically detect "people of interest" as they approach a campus and respond accordingly through access control, video surveillance, mobile applications and mass notification.

Currently most traditional physical security and access control systems are role-based systems that do not adapt dynamically to risk or threatening situations as they emerge in real-time. These systems can allow the wrong people or high-risk students into schools when they shouldn’t be allowed in, or conversely, keep the right people, such as law enforcement and first responders, out when they’re needed the most.

Intelligent physical security and access control systems can adapt to situations as risks increase. For instance, normally outside authorities do not have access to school cameras and video surveillance systems, or the ability to enter a school without permission.

With an intelligent system, when an active shooter situation occurs, access to view the facility can be opened to law enforcement and first responders with a smart device and proof of identity. Police nearby can communicate to look at school cameras to get eyes on and assess situations, and gain access automatically to go inside a school and respond to active threats. Furthermore, in an elevated risk situation, intelligent access control systems can prevent students and teachers from going into more dangerous areas.

In the case of an emergency, intelligent systems have mobile muster apps that make it possible for students to say if they are safe and where they are. Additional access intelligence can direct students and teachers to safe zones and areas out of harm’s way. Emergency assist apps can also help students communicate with law enforcement and first responders in the event of an emergency improving time to action.

Modern physical security systems integrated with A.I. can recognize high-risk individuals before they enter buildings. In the case of any school shooting, we must ask, could this have been prevented or at least mitigated? The answer is yes.

A.I. integrated with modern physical security can determine if someone is high-risk and help monitor them throughout the day or stop them at the door before causing any harm. Schools can be alerted when a high-risk student is in the building, they can advise safety officers that there’s a suspect or student in the building that’s worrisome, watch over them throughout the day, initiate active shooter protocol, and get law enforcement officers or first responders on the scene and inside the building much faster.

Keys for More Intelligent and Proactive Security Systems

When considering modernizing physical security systems, campus administrators should look for comprehensive safety solutions that detect early warning signs, assess and manage threats, and offer communication and coordination for faster response times within the school walls and with outside authorities to mitigate threats, while keeping parents informed of student safety.

The following capabilities should be considered when modernizing a physical security system:

Integration platform to make sense of it all. Go beyond a centralized monitoring interface and move to true integration. Modern physical security systems should have the capabilities or interoperability built in to integrate with outside technologies and data input sources including gunshot detection alerts, access control, facial recognition, license plate recognition, mobile devices, object detection, mass notification, real-time location, video management and other outside data inputs and sources.

A.I. assistance. Automatically detects "persons of interest" along different pathways to violence, which triggers user-defined system responses. Modern systems should no longer rely solely on active monitoring to detect evolving threats or initiate emergency procedures.

Sharable mobile operations. Create and share a common operating picture with local law enforcement, simplifying coordination and increasing visibility when it matters most. Users can manage dynamic situations on the move in real-time.

Mobile upstander alerts. The “See Something, Say Something” app empowers staff to centrally manage suspicions, as well as initiate emergency response from mobile devices. These technologies transform staff from bystanders into upstanders, so nothing slips between the cracks.

Mobile muster notifications. This capability helps ensure student safety. Once a staff member initiates an active shooter alert, push notifications arrive on each student’s mobile device prompting a simple “Safe/Not Safe” response. Each student response pushes their GPS coordinates to operators and parents, ensuring quicker response, while providing peace of mind.

A Painless Path to a More Prepared, Safe Campus

Implementing and integrating A.I. into physical security systems does not require starting over or complete ‘"rip and replace" overhauls. In many cases, A.I. and next-gen access control systems can augment existing security and access control systems, and be configured according to school district risks, privacy concerns and needs.

School violence has too frequently been in the headlines. Fortunately, legislation and technology are coming together in new ways to give schools the funding and tools they need to proactively fight back, arming everyone, with as much as possible, to improve the life safety of our students, children and teachers.

This article originally appeared in the May/June 2019 issue of Campus Security Today.