Netskope and Google Cloud Unveil AI Guardrails for Secure AI Workflows
Netskope, a leader in cutting-edge security and networking technologies for the cloud and AI era, announced a significant expansion of its partnership with Google Cloud on April 22, 2026, aimed at delivering robust security and performance for AI workflows. With this new solution, businesses can utilize Netskope One AI Guardrails to implement high-performance generative AI and autonomous agent workflows on Google Cloud at scale. By tapping into Netskope's capabilities, this solution enhances the safety of data utilized in these AI workflows.
As organizations are projected to invest over $867.3 billion globally in AI by 2029, a transition from simple chatbots to autonomous AI agents is becoming prominent. Consequently, there is a growing necessity for real-time protective measures specifically designed for these agents. The Netskope One AI Guardrails, as part of the Netskope One AI Security suite, ensures data protection against AI-specific threats and supports responsible use of AI through content moderation capabilities.
Utilizing Google Cloud’s Tensor Processing Units (TPUs), Netskope enables industry-leading cost efficiency and high throughput inference infrastructure that provides low-latency security even for the fastest AI workloads. By leveraging this solution, organizations are able to conduct rapid safety checks on optimized TPU instances. Netskope performs inline processing tailored to the processing speeds of the models, achieving extremely low latency without compromising performance.
With real-time moderation provided by Google Cloud's Vertex AI platform, Netskope One AI Guardrails can exploit a dedicated accelerator architecture that supports AI services on Google Cloud.
The features within Netskope One safeguard autonomous agents, ensuring that every single step they perform—whether calling tools, communicating with MCP servers, or executing autonomous tasks—is validated against enterprise policies in real time.
This solution is equipped to detect and block unintended recursive loops and malicious "ghost commands" that could potentially lead to resource depletion or unauthorized system changes on Google Cloud.
Moreover, real-time safety checks are conducted across various categories, including hate speech, discrimination, criminal activity, self-harm, piracy, copyright infringements, and requests for confidential information.
While traditional security measures focus on stored data, Netskope actively identifies and blocks AI-specific threats, such as prompt injection and jailbreaking, before the integrity of the models is compromised.
As the use of MCPs, APIs, and other tools by agents continues to grow, Netskope examines the data exchanged between agents and tools, preventing indirect prompt injections and unauthorized executions. All detection results are mapped to industry-standard frameworks like MITRE ATLAS™ and the OWASP Top 10 for LLMs, offering security teams a clear and auditable overview of associated risks.
Since Netskope's solution is directly integrated into customers' Google Cloud instances, confidential prompts and responses are scanned locally, ensuring that data remains protected without flowing outside the managed tenant. By applying consistent policies across all AI communications, customers can more easily comply with stringent regulations, including GDPR, HIPAA, and the EU AI Act.
Vineet Bhan, Director of Security and Identity Partnerships at Google, remarked, "The collaboration between Netskope and Google Cloud maximizes the strengths of Google Cloud's AI innovations alongside Netskope's enterprise security capabilities. Netskope One AI Guardrails seamlessly integrates with our Vertex AI and TPUs to facilitate the safe implementation of both generative AI and autonomous agents. Together, we're supporting businesses in building their future on a foundation of trust."
Netskope's CEO and co-founder, Sanjay Beri, added, "As a global leader in AI-centric security and networking technologies, Netskope provides essential functionalities for safely and efficiently deploying the latest AI workflows. We take pride in partnering with Google Cloud to empower our customers in confidently pursuing their AI utilization objectives."
Netskope's products are available via the Google Cloud Marketplace. This press release is based on the official announcement made on April 22, 2026, in the U.S. *1IDC Worldwide Artificial Intelligence IT Spending Forecast, 2025–2029 (August 2025, IDC #US53688725).
Netskope (NASDAQ: NTSK) is a leader in the security and networking technology of the cloud and AI era. The company delivers optimized access and real-time context-based security to an entire ecosystem that includes people, devices, data, autonomous agents, applications, tools, and large language models (LLM). Thousands of customers, including over 30 Fortune 100 companies, trust and utilize the Netskope One platform, Zero Trust engine, and powerful NewEdge network. These solutions visualize usage across all cloud, AI, SaaS, web, and private applications, helping reduce security risks and enhance network performance. For more information, visit netskope.com/jp.
As organizations are projected to invest over $867.3 billion globally in AI by 2029, a transition from simple chatbots to autonomous AI agents is becoming prominent. Consequently, there is a growing necessity for real-time protective measures specifically designed for these agents. The Netskope One AI Guardrails, as part of the Netskope One AI Security suite, ensures data protection against AI-specific threats and supports responsible use of AI through content moderation capabilities.
Achieving Responsible AI through Hardware Acceleration
Utilizing Google Cloud’s Tensor Processing Units (TPUs), Netskope enables industry-leading cost efficiency and high throughput inference infrastructure that provides low-latency security even for the fastest AI workloads. By leveraging this solution, organizations are able to conduct rapid safety checks on optimized TPU instances. Netskope performs inline processing tailored to the processing speeds of the models, achieving extremely low latency without compromising performance.
With real-time moderation provided by Google Cloud's Vertex AI platform, Netskope One AI Guardrails can exploit a dedicated accelerator architecture that supports AI services on Google Cloud.
Creating a Safe Ecosystem for Agent-based AI and Generative AI
The features within Netskope One safeguard autonomous agents, ensuring that every single step they perform—whether calling tools, communicating with MCP servers, or executing autonomous tasks—is validated against enterprise policies in real time.
This solution is equipped to detect and block unintended recursive loops and malicious "ghost commands" that could potentially lead to resource depletion or unauthorized system changes on Google Cloud.
Moreover, real-time safety checks are conducted across various categories, including hate speech, discrimination, criminal activity, self-harm, piracy, copyright infringements, and requests for confidential information.
Real-time Behavioral Protection Against Multi-vector AI Threats
While traditional security measures focus on stored data, Netskope actively identifies and blocks AI-specific threats, such as prompt injection and jailbreaking, before the integrity of the models is compromised.
As the use of MCPs, APIs, and other tools by agents continues to grow, Netskope examines the data exchanged between agents and tools, preventing indirect prompt injections and unauthorized executions. All detection results are mapped to industry-standard frameworks like MITRE ATLAS™ and the OWASP Top 10 for LLMs, offering security teams a clear and auditable overview of associated risks.
Native Data Sovereignty and Compliance Support
Since Netskope's solution is directly integrated into customers' Google Cloud instances, confidential prompts and responses are scanned locally, ensuring that data remains protected without flowing outside the managed tenant. By applying consistent policies across all AI communications, customers can more easily comply with stringent regulations, including GDPR, HIPAA, and the EU AI Act.
Vineet Bhan, Director of Security and Identity Partnerships at Google, remarked, "The collaboration between Netskope and Google Cloud maximizes the strengths of Google Cloud's AI innovations alongside Netskope's enterprise security capabilities. Netskope One AI Guardrails seamlessly integrates with our Vertex AI and TPUs to facilitate the safe implementation of both generative AI and autonomous agents. Together, we're supporting businesses in building their future on a foundation of trust."
Netskope's CEO and co-founder, Sanjay Beri, added, "As a global leader in AI-centric security and networking technologies, Netskope provides essential functionalities for safely and efficiently deploying the latest AI workflows. We take pride in partnering with Google Cloud to empower our customers in confidently pursuing their AI utilization objectives."
Netskope's products are available via the Google Cloud Marketplace. This press release is based on the official announcement made on April 22, 2026, in the U.S. *1IDC Worldwide Artificial Intelligence IT Spending Forecast, 2025–2029 (August 2025, IDC #US53688725).
About Netskope
Netskope (NASDAQ: NTSK) is a leader in the security and networking technology of the cloud and AI era. The company delivers optimized access and real-time context-based security to an entire ecosystem that includes people, devices, data, autonomous agents, applications, tools, and large language models (LLM). Thousands of customers, including over 30 Fortune 100 companies, trust and utilize the Netskope One platform, Zero Trust engine, and powerful NewEdge network. These solutions visualize usage across all cloud, AI, SaaS, web, and private applications, helping reduce security risks and enhance network performance. For more information, visit netskope.com/jp.
Topics Consumer Technology）
【About Using Articles】
You can freely use the title and article content by linking to the page where the article is posted.
※ Images cannot be used.
【About Links】
Links are free to use.