TLDR: Zenity has announced an enhanced integration with Microsoft Copilot Studio, providing enterprises with an end-to-end security and governance platform for AI agents. This collaboration aims to offer continuous visibility, vulnerability assessment, and real-time threat detection, enabling organizations to confidently scale their AI agent adoption while mitigating risks like prompt injection and data leakage. The partnership combines Microsoft’s robust platform for building copilots with Zenity’s agent-centric security approach, ensuring protection from development through deployment and ongoing usage.
Tel Aviv, Israel – Zenity, a leading provider of end-to-end security and governance solutions for AI Agents, has significantly enhanced its integration with Microsoft Copilot Studio. This strategic move is designed to empower enterprises to securely adopt and scale AI agents across their operations, addressing the growing complexities and inherent risks associated with advanced AI deployments.
As AI agents become increasingly sophisticated and embedded in critical business processes, new security challenges emerge concerning access controls, decision-making autonomy, and sensitive data handling. The enhanced collaboration between Zenity and Microsoft Copilot Studio aims to provide a comprehensive solution, ensuring that organizations can leverage the full potential of AI agents without compromising security or compliance.
Microsoft Copilot Studio offers an enterprise-grade platform that facilitates the rapid design and deployment of custom AI agents using natural language, logic flows, and pre-built connectors within the Microsoft 365 ecosystem. While Microsoft provides a strong foundation of governance and controls during the buildtime phase, Zenity extends this protection with a crucial layer of runtime security.
Zenity’s platform delivers continuous visibility, vulnerability assessment, and real-time threat detection across the entire AI agent lifecycle – from development (buildtime) through deployment and ongoing operational usage (runtime). This agent-centric approach focuses on understanding each agent’s intended function, its behavior patterns, and the data and systems it interacts with. Key features of this enhanced integration include:
Comprehensive Security from Buildtime to Runtime: Zenity enforces security guardrails during development, ensuring agents adhere to best practices and compliance standards, and provides continuous monitoring and threat detection during operation.
Policy Enforcement Across the AI Agent Lifecycle: Pre-deployment controls prevent vulnerabilities, overly broad access, and data exposure, while ongoing monitoring ensures agents operate within defined security parameters.
Real-Time Threat Detection and Response: Zenity’s AI Detection & Response (AIDR) system identifies risks like prompt injection, data leakage, and behavioral anomalies in near real-time, enabling automated responses based on predefined playbooks.
Full Observability and Transparency: Detailed activity logs and behavioral analytics offer deep insights into agent activity, usage, and access, explaining ‘what,’ ‘why,’ ‘by whom,’ and ‘with what data.’
Harrison Johnson, VP of Partners and Alliances at Zenity, emphasized the importance of this integration, stating, “Zenity enables enterprises to innovate with AI Agents safely and confidently by ensuring that every agent is governed, secured, and observable from buildtime to runtime.” This sentiment is echoed by the broader industry, with Forrester recognizing Zenity in its ‘AI Governance Solutions Landscape, Q2 2025,’ highlighting its leadership in securing and governing AI initiatives.
The partnership addresses the critical need for robust security programs for AI agents, akin to insider risk programs but tailored for AI. Michael Bargury, CTO and co-founder of Zenity, noted that enterprises are adopting AI at an unprecedented pace, underscoring the necessity of trust in AI agents for production systems. Zenity’s platform helps discover and remediate risky agents, often with automated fixes.
This announcement comes at a time when AI agent security is paramount. Zenity Labs recently unveiled ‘AgentFlayer’ vulnerabilities at Black Hat USA 2025, demonstrating 0-click exploit chains that could silently hijack major enterprise AI agents from vendors including OpenAI, Microsoft, Salesforce, and Google. These findings underscore the urgent need for the proactive and comprehensive security measures that Zenity’s enhanced integration with Microsoft Copilot Studio provides, allowing enterprises to scale AI innovation with confidence and control.
Also Read:
- Concentric AI Enhances Enterprise Data Security with OpenAI ChatGPT Integration
- Singaporean Firms Excel in AI Security Amidst Rising ‘Shadow AI’ Challenges
Ultimately, this collaboration empowers platform and security teams to accelerate AI adoption without sacrificing control, fostering secure and scalable innovation by design.
