Demisto Applies Machine Learning Algorithms to IT Security Operations

Most security professionals spend a lot more time looking for the source of problem than they do fixing it. That can be especially vexing when they know there’s a cyberattack in progress. To cut down on that time, Demisto this week added a Demisto Insights module to a Demisto Enterprise security operations platform that provides access to machine learning algorithms that can now suggest the best method available for resolving a security issue.

Rishi Bhargava, vice president of marketing for Demisto, says the two biggest security issues IT organizations face when it comes to personnel are a shortage of staff and the amount of time it takes for anyone to gain a relevant amount of IT security expertise.

“They can use Demisto Insights to set up playbooks for dealing with certain types of threats,” says Bhargava.

Demisto Insights is not only designed to make it possible for IT security professionals to spend less time diagnosing problems; Bhargava says it provides the added side benefit of making less experienced security personnel staffing a security operations center (SOC) become more effective sooner. That module plugs into a security information event management (SIEM) platform that Demisto wrote from the ground up to correlate incidents in a way that makes it easier to automate IT security responses, says Bhargava.

As IT security technologies become imbued with advanced analytics and automation technologies based on machine learning algorithms, it’s not clear how much longer a chronic shortage of IT security staff will persist. In the meantime, most existing security professionals would prefer to spend their time hunting for sophisticated malware versus dealing with the same routine issues time and again. As such, there’s a general expectation that security professionals will be less resistant to these advances than other IT professionals focused on other aspects of IT operations.

Regardless of how anyone may feel about it, however, the machine learning genie is out of the bottle. The only real issue now is determining where the capabilities of the machine end and the value a human brings to the equation begins.