Search
Close this search box.
Search
Close this search box.

ForgePoint Capital’s William Lin : A Conversation on Cybersecurity

ForgePoint Capital’s William Lin

RR: Can you give us a general view of Machine Learning (ML) in the Cyber world?

Lin: Broadly, ML models are fairly mature nowadays. And can be used for quite a few security use cases!

RR: What ML models do you see outperforming others?

Lin: Deeper, attention-based models have made big leaps in machine learning. Classifying data objects at a cloud-scale is a natural use case that powers many incidents response and compliance workflows. Symmetry Systems is a portfolio company building the Cloud Data Security category. DeepSee is another portfolio company building the Knowledge Process Automation category.
In both cases, they apply Machine Learning models to help build classifiers and knowledge graphs that were impossible before.

RR: What ML models are you excited about in the future?

Lin: Leveraging new advancements in NLP/NLU will provide a step function improvement in Identification and Prevention security use cases. Organizations and their customers create so much data, NLP/NLU is invaluable in helping a company understand where a company’s riskiest data is, how it is flowing throughout the organization, and in building controls to prevent misuse.

How about for an ML use case scenario, would you mind giving us one as an example?

Lin: For Detection use cases, we can look at NLP/NLUs learnings used in spam and email security. While more sophisticated models have made spam harder, attackers have also used similar models to generate more sophisticated spam. This cat and mouse game usually reaches equilibrium before the detection models filter out good emails alongside highly sophisticated spam.

RR: Can you tell us a little about how ML is used to defend against hacking attacks?

Lin: Attackers start with figuring out how to get into a company first while defense has to figure out what they have first. Hence why we see so many security companies that filter, prioritize, detect and automate workload early in category creation. NLP/NLU is especially well suited to help defenders figure out what they have.

RR: So will NLP be the Cybersecurity ML winner in time?

Lin: There is a lot of work ahead for the industry for NLP/NLU to become commonplace, defenders should have the home field advantage and we intend to help make that a reality with our investments.

Interview by Göktuğ Önyer

ForgePoint Capital’s William Lin