Zscaler’s ThreatLabz report: AI growth and threats – SiliconANGLE News

Zscaler Inc. has released its inaugural “ThreatLabz 2024 AI Security Report,” and the results document explosive growth in artificial intelligence, a concurrent rise in blocked AI traffic and a growing threat landscape as bad actors increasingly leverage the technology.

The report drew from over 18 billion transactions in the Zscaler Zero Trust Exchange between April 2023 and January 2024. The report’s findings highlight the need for enterprises to securely enable AI as global adoption continues to rise.

“It’s a transformative time for cybersecurity as a whole,” said Deepen Desai, chief security officer of Zscaler. “With the advent of generative AI, there is a huge opportunity where it comes to applying both generative and predictive models in transforming how security operations are done. We also have to be wary of the fact that the bad guys are also going to be leveraging this and targeting your organization. It’s already happening.”

Desai spoke with Shelly Kramer, managing director and principal analyst with theCUBE Research, about key findings from the Zscaler analysis.

Report from Zscaler notes growth in usage and blockage

Zscaler’s report found explosive AI growth, with enterprise AI/machine learning transactions surging 595% over the last three quarters in 2023.

“The data makes it very clear that AI is no longer just an innovative thing, it’s actually starting to become business as usual,” Desai said. “It’s across several different industries where we’re seeing this adoption. That trend will continue as more organizations adopt these applications.”

The report also noted that even though AI usage has accelerated dramatically, enterprises are blocking 18.5% of all AI transactions, an increase of 577%. This rise in blocked transactions reflects concern around unsanctioned or “shadow” AI use inside an organization.

“That’s a natural reaction when there are a lot of unknowns around new technology and the risks that it brings,” Desai said. “You would start with that big hammer approach and as you understand more, you can open things up.”

The report’s results highlighted an emerging AI threat landscape as malicious actors seek to leverage the technology. This includes AI-driven phishing campaigns, social engineering attacks and the use of increasingly more sophisticated deepfake tools, according to Desai.

“As you enable organizations to adopt AI securely, it is still adding to your attack surface,” he explained. “You see very, very persuasive phishing pages, phishing emails being crafted using AI. You need to have zero trust. It’s only going to keep getting worse.”

In fact, Desai asked Zscaler’s security research arm, ThreatLabz, to simulate an attack using ChatGPT.

“With careful prompt engineering, they were able to have it [generate] phishing email, phishing pages that looked very similar to the real pages out there,” Desai said. “There are many more [variants] that keep popping up that are already available as a service for the bad guys to use.”

What can enterprises do in the face of growing AI use and an increased threat landscape? Desai advises a process that screens AI applications and closely examines the security posture of private large language model environments.

“Policies and effective control will be the difference between doing this thing the right way versus the wrong way,” Desai said. “Most organizations are working on some sort of private LLM infrastructure. You have to have a plan in place to protect that environment. This is where we’re going to see more adversarial attacks happening.”

Here is the complete video interview, and be sure to check out more of theCUBE’s CUBE Conversations:

[embedded content]

Image: Getty Images

This post was originally published on 3rd party site mentioned in the title of this site