Cisco CEO Chuck Robbins explained to ZDNet the importance of AI in enabling 'every single piece of technology that we build', particularly across security offerings that were able to detect VPNFilter as a result.
Artificial intelligence (AI) is an enabler of technology rather than being a business itself, according to Cisco CEO Chuck Robbins, who told ZDNet that AI is critical in sitting behind much of the networking giant's portfolio.
"Across our portfolio ... if you look at how we deal with security and the number of threats, there's an element of AI and machine learning, and we wouldn't be able to do what we're doing [without it]," Robbins told ZDNet.
"When our threat researchers found the VPNFilter attack, there was a lot of information and a lot of compute power going into discovering these things on a global basis, so it's important there, it's important in intent-based networking, it's important in our collaboration portfolio, it's important everywhere.
"Some companies will decide that AI is a business for them. For us, it is an enabler of every piece of technology that we build."
Robbins added that while many companies are mislabelling things as AI, Cisco is more "realistic" about its actual definition.
"We're realists about what technology is really doing, and I personally think a lot of what is being called AI today is simply massive datasets with incredibly intelligent algorithms being processed very quickly," he argued.
Calling AI and machine learning "pervasive" across the business, Cisco's EVP of Networking and Security David Goeckeler also pointed to Talos' discovery of VPNFilter, attributing this to AI.
"That's a very good example where with our threat intel research, we actually then reach out to governments and we coordinate activities of how to protect people from cyber threats," he told media at Cisco Live.
"So there's an enormous amount [of AI] that goes on across the board on cybersecurity."
Goeckeler boiled down Cisco's approach to AI as being an "enormous dataset that we apply machine and human intelligence to find out where threat actors are in the world, and we push policy back into the infrastructure".
Cisco is particularly focused on utilising AI and machine learning across its security portfolio, he said, and has been for "a very long time now".
"We're streaming real-time telemetry to a central point in the network, and then we're applying intelligence, we have recommendation engines, when we see something we'll recommend what the fix should be, so all of that is live across the networking portfolio today as well, and then you've got the intersection of these too, like ... encrypted traffic analytics, so essentially taking networking data plus some security data and you're mixing them together and you're using inference to figure out what is malware," Goeckeler explained.
"You can't inspect it anymore because it's encrypted, but if we collect enough data of the behaviour of traffic, we can infer with very, very high probability and very, very low false positives what is malware, so that's a perfect example of where we're using very, very advanced learning techniques kind of intersected with the datasets we have both on the network data and the security data."
"I think right now what Amy [Chang] brings is a real deep analytics, AI, software expertise, and I think if you look at where we're going to -- how we're going to work that portfolio in the future -- it really is going to be around bringing in intelligence to create more robust experiences for those that are using our platforms," Robbins said.
Disclosure: Corinne Reichert travelled to Cisco Live in Orlando as a guest of Cisco