At GSX 2022, Vector Flow convened an expert panel to discuss Data-Driven Security together with:

  • Vik Ghai, CTO, Vector Flow, Inc.
  • Daniel Tsybulsky, CPP, Associate Director, Corporate Security at Gilead Sciences
  • Loren Hansen, CPP, CISSP, Safety/ Security Technology Manager at Juniper Networks
  • Ed Chandler, Chairman, Security By Design

They covered a range of issues and provided their advice on a few ways to streamline and improve the implementation and digital transformation process.

Here’s a recap of some of the topics and comments.

Reactive Security Challenges

The availability of operational data, and the growing AI-powered ability to collect, ingest, process, and draw conclusions from that data, is driving new intelligence into many organizational processes.

Unfortunately, in too many organizations, Security remains challenged to sift through enormous quantities of data without AI tools and thus cannot gain the benefits that are possible. Without the ability to find the important conclusions in real time, too many have remained reactive, rather than data-driven and proactive.

“It has been a desire of the security industry for the last 25 years to become more proactive rather than reactive, and the data allows us to do that now.”   – Vik Ghai, CTO at Vector Flow

Benefits of Data-Driven Technology

data-driven approach visual

The team briefly covered some of the benefits that have been highlighted by Booz Allen Hamilton, including the “first-order” benefits of an overall greater level of real-time visibility of organizational situations and the transition to more proactive security. Secondary benefits include cost savings, business resiliency, and improved customer understanding and intimacy.

Data-Driven Technology Adoption in Daily Life

Society, at large, has been increasing the demand for, and use of, data-driven automation in our daily lives, the use of Maps, Google Maps or Uber are typical examples. From “reactive” approaches looking at paper maps or directions, moving toward “predictive” data uses such as Google Maps that can predict the estimated times of different routes to your destination. With the increasing power of AI engines, such programs can now go one step further, to “prescriptive” recommendations that provide the fastest or best path, optimized with all the factors included, to achieve the goal.

Similar changes are happening in security-related systems, including video surveillance systems, access control systems, intrusion detection, and so on. Vik Ghai noted, “pulling data from these systems can provide a basis for predictions about what will actually happen.” In this way, systems are moving from describing what happened in the past (reactive mode) to making predictions about the future. AI engines such as Vector Flow can now make recommendations about, for example, what access to provide to a new hire based on the actual usage data of peer workers in the same department or role.

Mr. Hansen related several benefits he’s already seen by implementing a data-driven approach, including detecting behaviors that suggested lost badges, duplicated badges, and even the possession of an unauthorized master key.

“By using data-driven automation, we have reduced risks and also improved operations. For example, automatically, and dynamically, changing parameters to account for the need for longer door open times during lunch (to carry food) reduced nuisance alarms and refocused security operators on more critical tasks.”  – Mr. Hansen

Implementation Challenges

There are two challenges, The first is understanding the value and the meaning of the data. The second is the lack of synergy between departments and their needs to improve security and processes.

“I think you have to start with a commitment to data-driven decision making,” said Mr. Tsybulsky. “It may be easier at a science company, where data is a more in its culture. But you can’t really understand how to fix security performance at any organization until you have the data.” 

Mr. Hansen related that one challenge is that “the data is already there. It’s always been there. But getting the insights and information from it, in time to do something, is a challenge. And some manufacturers seem to have overlooked important data too – like indicating that a device might need repair. Fortunately looking at other indicators can sometimes reveal these insights.”

Several panelists noted that once other departments became aware that Security had this valuable data, they came running with new requests for insights that would help them do their jobs as well.

How to Streamline Digital Transformation

The panelists covered that the industry is waking up to the fact that we have two primary sources of big data – access control and video surveillance. We have all this data – now we can use it to help us make better decisions.

To justify the budget for data automation, Mr. Tsybulsky added, “We started with a simple calculation – how much would it cost for us to collect and interpret this data manually? Then, we added the value of removing the manual work and letting operators focus on higher-value tasks.”  He added that acquisitions are another challenge because of limitations on SOC stations. Employing data automation that scales is another way to help stay on budget even as the number of monitored system elements increases.

Mr. Hansen noted that reducing the effort required to address false and nuisance alarms may alone justify the entire cost of the system. He also added this ‘heads-up’: “One surprise was that as we implemented data automation, one of the first things that happened was that we found out how many things needed repair – so we had more work! But that’s not a bad thing – taking care of those things greatly improved our security stance.”

Contact us to learn more about how Vector Flow can help you move toward a data-driven security environment that makes your workplaces safer and more efficient.

By: Vector Flow