Cybersecurity in the era of artificial intelligence is not just about following regulations, it’s about making compliance and resilience intrinsic to AI strategy.

Following that ethos, VMware by Broadcom has added services to VMware Cloud Foundation 9. The goal is to keep private AI models secure, governed and capable of fast recovery.

Broadcom’s Vijay Ramachandran discusses cybersecurity for AI.

“We have to provide a set of tools as VMware for our customers to prove to their auditors that they’re secure,” said Vijay Ramachandran (pictured), head of product management and core infrastructure, VCF Division, at Broadcom Inc. “Cybersecurity or cyber resilience is all about how do you do it? What are the mechanisms and what are the tools to protect your data? It’s not just protecting your data, it’s protecting workloads.”

Ramachandran spoke with theCUBE’s John Furrier and Rebecca Knight at VMware Explore, during an exclusive broadcast on theCUBE, SiliconANGLE Media’s livestreaming studio. They discussed VMware’s strategy for cyber resilience and the impact of AI on cybersecurity. (* Disclosure below.)

The roadblocks in modern compliance

Protecting AI, particularly private models, has proven itself to be a complex task. Obstacles include the increasingly wide distribution of data in modern organizations and changing regulatory standards.

“Regulators used to come once a quarter,” said Ramachandran. “Now, customers are expecting continuous compliance checks. We need to provide a platform that does continuous compliance checks and remediation. Things have changed. Things are a lot more spread out … and you need to be a lot more in real time.”

VMware has extended endpoint detection and recovery — typically used for devices such as laptops or phones — to hypervisors, providing broader protection for its customers. Equally as important, according to Ramachandran, is ensuring a strong recovery should the worst case scenario happen. In the age of AI, anything is possible.

“Customers have lost data because they entered the wrong prompt,” he said. “You’re going to see this happen a lot more. And especially with AI, bad actors can actually use that to wipe out an entire real estate of infrastructure. At VMware, our goal is to provide tools to protect against not just the old threat vectors, but the new ones as well.”

Here’s the complete video interview, part of SiliconANGLE’s and theCUBE’s coverage of VMware Explore:

(* Disclosure: Broadcom Inc. sponsored this segment of theCUBE. Neither Broadcom nor other sponsors have editorial control over content on theCUBE or SiliconANGLE.)

Photo: SiliconANGLE