Netskope has introduced Netskope One AI Security, a set of AI security features that protect, make visible and accelerate the entire AI ecosystem, unified within the Netskope One platform.
Netskope One AI Security introduces new products to drive and enable the AI era, along with AI Index, the first index of its kind, providing global tracking of AI adoption and new risks.
These are the new products:
• Netskope One Agentic Broker: offers visibility and control over all MCP (Model Context Protocol) transactions, both authorized and unauthorized, allowing organizations to secure interactions with AI agents.
• Netskope One AI Guardrails: Prevents AI threats such as command injection and jailbreaking, and offers LLM content moderation to enable responsible use of AI.
• Netskope One AI Gateway: inspects and applies security policies to private AI and/or LLM applications that organizations host or develop themselves.
• Netskope One AI Network Teaming: simulates hostile attacks, detects weaknesses, and assesses security risks of LLMs and AI applications.
Business investment in AI continues to grow at a rapid pace: according to IDC, global spending on AI would reach $241.8 billion in 2025 and is projected to exceed $867.3 billion in 2029.
This rapid adoption of AI has also created significant security gaps, as organizations using legacy security products lack visibility into how their data is used across AI-based applications, private and public AI models, and AI agents and tools. This is because AI systems often bypass security and introduce risks such as data leaks, manipulative prompts, and misuse.
Ecosystem within a unified and integrated platform
Netskope One AI Security allows you to know, view and control in real time all the AI applications, models, agents and tools that are being used, analyze their specific risks and accelerate the safe adoption of AI throughout the ecosystem within a fully unified and integrated platform. Additionally, customers benefit from Netskope NewEdge AI Fast Path, a set of features that efficiently optimize latency to their AI destinations.
Combining the high-performance access of NewEdge AI Fast Path with the intelligent, contextual zero-trust controls of Netskope One AI Security provides fully optimized and integrated performance, resilience, and security for enterprise AI.
“The AI supercycle is here and demands a new standard of security and high-performance network connections. We believe that the next decade will be marked by an intelligent perimeter. Our Netskope One platform, combined with our NewEdge network, is precisely that: an architecture built specifically to meet the requirements of an autonomous and agency economy.
This architecture provides a native AI foundation with the intrinsic ability to protect the complex data flows that underpin the modern AI ecosystem,” said Sanjay Beri, co-founder and CEO of Netskope. “With the launch of our new AI security products and our AI Fast Path infrastructure, we deliver in-depth, real-time protection at the speed of inference. “By unifying high-speed performance with the active context of billions of transactions, Netskope provides the adaptive framework essential for the modern AI enterprise.”
Alexander Schuchman, CISO and Senior Vice President of Global Network Engineering and Operations at Colgate-Palmolive Company, said: “Colgate-Palmolive’s 220-year history of innovation is the foundation of our company’s purpose as an innovative, caring company that is redefining a healthier future for all.
We leverage technology solutions, including artificial intelligence, to drive creativity, efficiency and reach. AI systems offer powerful tools to improve our business processes, our customer and consumer experiences, and innovation across our operations.
However, technology comes with responsibilities, which is why we are guided by our Employee Code of Conduct and our AI governance structure. We ensure the responsible deployment and use of AI systems, prioritizing good judgment, compliance with applicable laws and regulations, and ethical considerations, in order to mitigate any unintended consequences. “Colgate-Palmolive continues to collaborate with Netskope to provide data protection associated with our AI systems and related use cases.”
What’s new with Netskope One Agentic Broker
● Provides built-in real-time visibility and protection for the autonomous AI ecosystem by decoding and protecting Model Context Protocol (MCP) traffic between AI agents and data sources.
● Provides real-time monitoring and control over how AI agents interact with enterprise data sources.
● Ensures a consistent security posture that protects sensitive company data and enables rapid, large-scale agent automation.
What’s new in Netskope One AI Guardrails
● Protects against new AI-based attacks, such as jailbreaking and command injection, and acts as a real-time content moderator for every user and automated agent interaction.
● Seamlessly integrates with Netskope One DLP and Threat Protection into a single cohesive view, allowing security teams to understand the context of detections without having to manage fragmented alerts.
● Automatically detects and blocks discriminatory, copyrighted and inappropriate content, which can pose a legal and reputational risk to the company.
● Map detections to MITER ATLAS and OWASP Top 10 to keep LLMs ahead of adversary tactics.
This architecture provides a native AI foundation with the intrinsic ability to protect the complex data flows that underpin the modern AI ecosystem.
What’s new with Netskope One AI Gateway
● Extends Netskope AI security controls to protect agent communications in private AI environments (on-premises or virtual private cloud) that do not pass through the cloud for inspection, and centralizes authentication, traffic management, and content inspection.
● Enables organizations to secure interactions between applications and LLM, and ensure that agents’ autonomous data flows remain controlled and secure.
What’s new in Netskope One AI Red Teaming
● Extends Netskope protection to the development cycle to proactively prevent and eliminate vulnerabilities in private AI deployments.
● Expose AI models to thousands of simulated attacks to identify where complex, multi-round attacks could bypass the AI model’s default guardrails and stop performance drift.
