SAP CSO: Security Requires Context
Security depends on the apps and networks it protects. SAP CSO Justin Somaini discusses three scenarios.
June 11, 2018
When Justin Somaini talks about security, "context" comes up as often as any technology or process.
"How do we think about security? How do you bring the context of security to the application context?" he asked when Dark Reading interviewed him at the recent Sapphire NOW conference in Orlando, Fla.
Pivoting to an answer, he admitted that the answers are easier for some applications than others. He went on to discuss a trio of context types.
1. Critical Contexts
In an application such as Concur, SAP's travel expense management suite, "there's a lot of travel security in there. There's also fraud protection," Somaini explained. The challenge, he said, is identifying the same sort of contextual framework for other applications, such as those supporting HR or the supply chain.
But doing so is critical. "For us, being able to take security into the line-of-business context is critical," Somiani said. "It's where security really meets the business."
An increased use of analytics is one of the ways Somaini is working to find context. The questions he works through are similar to those many have in data-heavy security: "How much [data] do you need? What are the signals going in that return useful information?" he asked.
The answer is a work in progress. "I believe there's more we can do, but we haven't really figured it out," he admitted.
2. The IoT Context
Talk turned to the Internet of things (IoT). In that context, Somaini said, a particular security difficulty comes with the long reach of the devices at the edge.
"In the IoT, you're including a dependency on the supply chain," he explained. In other words because so many IoT devices are difficult or impossible to upgrade or modify, to a big extent customers are reliant on the security decisions made by device manufacturers.
In some cases, however, IoT can be easier to secure because the machine-to-machine communications that make up so much of the IoT aren't dependent on the vagaries of user interaction, Somiani said. Yet at the same time, he added, the sheer breadth of the IoT brings its own set of challenges.
"You have to trap the big network that might include cellphones or robots on the factory floor or rail transport," he said. That huge span provides a great deal of room in which criminal activity can hide. For example, a malicious actor from outside of the organization might try to masquerade as someone on the factory floor or as a device on rolling stock.
That possibility means the context has shifted. "Everything we've done previously has been digital, but IoT takes it into the real world," Somaini said.
3. The Employee Context
As the interview neared its end, an SAP employee spotted Somaini and took the opportunity to ask a question about the FBI's recommendation of action on VPNFilter. Somaini answered, and then turned to larger questions about the responsibility of enterprise IT security in the face of such threats.
First, he admitted that it might well be time for CSOs to look at programs to either remediate problems on employees' home routers or show the employees how to remediate issues on their own devices. The issue, he said, is complexity.
"We made the configuration and management of systems incredibly painful. It's impossible for a nonsecurity professional or an end user to set things up," Somaini said.
Somaini also pointed out the similarities between the security challenges of the IoT and those of employees at home. Ultimately, he said, the answer will come in building more security into the systems that machines, enterprise employees, and consumers use to do their work.
Related Content:
Top industry experts will offer a range of information and insight on who the bad guys are – and why they might be targeting your enterprise. Click for more information
About the Author
You May Also Like
Applying the Principle of Least Privilege to the Cloud
Nov 18, 2024The Right Way to Use Artificial Intelligence and Machine Learning in Incident Response
Nov 20, 2024Safeguarding GitHub Data to Fuel Web Innovation
Nov 21, 2024The Unreasonable Effectiveness of Inside Out Attack Surface Management
Dec 4, 2024