Thales has entered into a strategic partnership with Google Cloud to assist organizations that are facing new security, privacy, and compliance challenges around agentic AI. The partnership enables continuous protection, visibility, and automated policy enforcement for workflows powered by agentic AI within the Google Cloud environment.
Thanks to the integration of Thales‘ AI Security Fabric with Google Cloud Gemini Enterprise, enterprises will be able to enforce security controls and governance policies on interactions that involve human users, agentic AI agents, foundation models, and connected corporate databases.
Agentic AI: Mitigating Risks throughout the Lifecycle
While modern enterprises have shifted from traditional conversational assistants to more sophisticated autonomous AI agents that can plan actions, execute tasks, and communicate with other systems, it has become evident that static security perimeters cannot provide adequate protection for such advanced systems. In contrast to legacy applications that work in a predictable and hard-coded manner, autonomous agents dynamically interact with each other and other components, creating an extended and highly fluid attack surface.
The protection of agentic AI workflows has been recognized by industry experts as one of the most important priorities for corporate IT teams today.
Also Read: Arcadis and Autodesk Expand Partnership to Scale AI-Driven Workflows Across Built and Natural Environments
The unified security framework addresses vulnerabilities unique to generative and agentic systems, including:
Prompt Injection Defense: Intercepting malicious inputs engineered to override core safety parameters or extract confidential system instructions.
Sensitive Data Leakage Prevention: Applying real-time inspection and data loss prevention (DLP) filters to halt unauthorized disclosure of personally identifiable information (PII) or trade secrets.
Agent Boundary Enforcement: Setting strict behavioral guardrails to prevent autonomous systems from exceeding administrative mandates or pulling unauthorized datasets during execution (e.g., preventing an insurance processing agent from utilizing unverified external consumer records during claims settlement).
Agent-to-Agent Governance: Auditing and securing inter-agent communications and API calls across multi-model environments.
“Enterprises are moving from AI assistants to AI agents that can autonomously take action, make decisions, and interact with critical business systems. This requires a fundamentally different approach to security,” Eva Rudin, Senior Vice President, Cybersecurity Products, at Thales, said. “By expanding our collaboration with Google Cloud, we are helping organizations build the necessary trust with security and governance as the foundation required for the next generation of enterprise AI.”
“Google Cloud helps enterprises move from experimenting with AI agents to deploying them securely across their businesses,” Vineet Bhan, Director of Security and Identity Partnerships, Google Cloud, said. “As agents become more capable and connected to critical data and systems, customers need security and governance built into how they operate. Our work with Thales gives organizations additional capabilities to protect agentic AI workflows and adopt this next generation of AI with greater confidence.”
Empowering Secure Enterprise AI Scale
By combining Google Cloud’s high-performance AI deployment platforms with Thales’ deep domain expertise in data security, identity access management, and runtime threat protection, the collaboration enables risk-sensitive organizations across finance, healthcare, government, and critical infrastructure to confidently scale autonomous agentic capabilities. Security leaders gain centralized auditability and policy enforcement, ensuring AI-driven business automation operates safely within organizational policy and regulatory mandates.