Zscaler acquires Splx to enhance AI app inspection and protection – News

Cloud security company Zscaler Inc. today announced its acquisition of SplxAI Inc., a security platform designed for artificial intelligence, in a bid to bolster the firm’s AI asset discovery and protection for enterprise teams.

Zscaler said it intends to expand the company’s zero-trust capabilities by adding Splx’s AI asset discovery, automated red teaming and governance to protect AI applications and services from development through deployment.

“Today marks an important step in advancing Zscaler’s role as the trusted partner helping organizations securely adopt AI,” said Jay Chaudhry, founder, chairman and chief executive of Zscaler. “AI is creating enormous value, but its full potential can only be realized when it can be secured.”

Splx launched AI Asset Management earlier this year, an extension to the company’s AI security platform that provides enterprise organizations with visibility into AI models, agentic workflows and infrastructure.

This capability is combined with threat inspection, AI prompt hardening and automated red teaming services. AI red teaming is the structured process of simulating attacks from malicious adversaries in an attempt to identify vulnerabilities and flaws in AI systems before real-world attackers can exploit them.

Red teaming is a fundamental part of maintaining and hardening the security of existing and newly developed software systems. Many AI models have exploitable vulnerabilities, which must be overcome by carefully filtering, curating and monitoring prompts sent to models as well as their outputs.

The Splx platform includes more than 5,000 purpose-built and domain-specific attack simulations designed to uncover risks and vulnerabilities. After issues are revealed, Splx will offer real-time suggestions on how to fix security issues before they become a problem.

The adoption of AI is driving record infrastructure investments. Total AI spending worldwide is forecast to total around $1.5 trillion in 2025, with infrastructure accounting for a significant part. Globally, companies are forecast to spend $375 billion in 2025 on AI infrastructure alone, according to a report from Virginia Tech, representing a 67% increase over last year.

“Zscaler and Splx share a vision to confront the vast new attack surface created by rapidly expanding AI infrastructure investments,” said Splx co-founder and CEO Kristian Kamber.

This attack surface encompasses various new and largely untested AI applications that utilize lengthy software chains to automate and accomplish tasks. They access both internal and external knowledge, and mostly operate with minimal or no human intervention.

New AI assets, such as AI agents, can act almost entirely autonomously, connect using model context protocol to other AI agents and data sources and are expanding to cover virtually every facet of software and data work across the enterprise.

Splx’s platform allows companies to map agentic workflows and visualize every node, agent and tool across complex software chains and decision-making processes. This provides development and security teams with the ability to stay ahead of how AI systems interact, identify dependencies and uncover potentially insecure connections. Agent-level threat analysis provides an additional layer for AI tools by delivering real-time, prioritized insights into risks.

Splx is a venture capital-funded startup with around $9 million in funding across two rounds, including a round of $7 million in March. Investors include LAUNCHub Ventures, Rain Capital, Inovo Runtime Ventures, DNV Ventures AS and South Central Ventures.

Zscaler said the new acquisition will be used to expand its existing capabilities to expand protection across the entire AI and data lifecycle on one platform. The addition of Splx follows the company’s acquisition of Red Canary Inc., an AI-driven threat management company focused on agentic AI, and an expansion of the company’s existing partnership with CrowdStrike Holdings Inc., a leading cloud-based cybersecurity company, to strengthen AI-enabled security operations.

Image: News/Microsoft Designer