BlueVoyant launches first dedicated AI agent security service for Microsoft environments. Jennifer Smith, Content Manager | Published 6 July 2026 The first security service built to govern AI agent identities, permissions, and behaviour before they become a liability BlueVoyant, a leading cybersecurity company, today announced the availability of its Microsoft Agent 365 Security Deployment Service, the first professional services program to help enterprises discover, govern, and secure the AI agents running across their Microsoft environment. The service launches as organisations worldwide race to adopt AI agents faster than their security teams can govern them. Enterprises are integrating AI agents (including Microsoft 365 Copilot, Copilot Studio, Azure AI Foundry, and third-party agents) into critical business workflows at an unprecedented pace. But these agents represent a fundamentally new class of non-human identity. These agents can read sensitive data, call external tools and MCP servers, and act autonomously on behalf of users and organisations. Most enterprises have no inventory of which agents exist, no identity or access policies applied to them, and limited ability to detect or respond to agent-related threats. "AI agents are the fastest-growing attack surface that most security teams aren't watching," said John Hernandez, CEO at BlueVoyant. "The question isn't whether enterprises are running agents. They are. The question is whether those agents are governed. Right now, for most organisations, the answer is no." A Control Plane for the Agentic Enterprise The BlueVoyant Microsoft Agent 365 Security Deployment Service addresses this gap by establishing Microsoft Agent 365 as a centralised control plane for AI agent governance. The engagement configures Agent 365 alongside the Microsoft security stack (Microsoft Entra, Microsoft Defender for AI, and Microsoft Purview) so that AI agents are governed with the same rigour as any other identity in the enterprise. Over the course of a guided, approximately 90-day engagement, BlueVoyant's experts work directly within the client's Microsoft tenant to deliver: * Agent inventory and registry: A view into which agents exist, who owns them, what data they can access, and what tools they can invoke - including shadow AI detection for agents deployed without IT knowledge. * Identity and access controls: Agent Conditional Access and Identity Protection policies configured in Microsoft Entra, treating agent authentication with the same scrutiny as human user authentication. * Threat detection and response: Microsoft Defender Security for AI enabled and integrated with Defender XDR, providing detection and response capabilities specific to agent-related threats. * Data protection policies: Microsoft Purview DLP and Insider Risk Management extended to AI agents, reducing exposure from data oversharing and unauthorised access. * Client-owned configuration: All configuration lives within the client's tenant. BlueVoyant delivers the expertise and the blueprint; the client retains full ownership at engagement close. Addressing a Market Gap with Microsoft The service is designed to attach directly to Microsoft 365 E7 and Agent 365 deployments and is aligned with AI Cloud Partner Program incentives. It positions BlueVoyant as one of the first security partners with a defined, repeatable AI-agent security offer - a strategic advantage as the agentic AI market matures. "At the conclusion of the engagement, clients will have clear answers to the foundational questions every security leader should be asking today: What AI agents do we have? Who owns them? How are they governed? And what happens when one is compromised" said Sebastian Sobolev, Chief Product Officer at BlueVoyant. The Microsoft Agent 365 Security Deployment Service is available immediately. Clients must have Microsoft E5 or E7 licensing with the Agent 365 add-on to be eligible.

The Buzz: Pax8 crowns the MIP era at Beyond26, Arrow launches partner experience centers, and Mitel names a new channel chief. Pax8's Salt Lake City conference signals the next business model shift for MSPs, while Arrow and Mitel make their own channel moves. * Pax8 Beyond26 - managed intelligence: Pax8 wrapped its annual Beyond conference in Salt Lake City on Tuesday with over 3,500 attendees including 200+ from Canada, centering the show on the transition from managed services to the Managed Intelligence Provider model. The headline announcement was Microsoft Agent 365 for Managed Intelligence - multi-tenant governance of agentic AI across MSP client environments through the Pax8 Agent Store, arriving in July - alongside the launch of the Managed Intelligence Provider Program, Voyager Alliance Rewards, and the Managed Intelligence Alliance. CEO Scott Chasin argued that as AI models commoditize, the trust MSPs have already built with clients is their primary competitive advantage going forward. * Arrow Electronics global experience centers: Arrow introduced a network of global experience centers on Tuesday, built in close collaboration with channel partners in North America and Europe to reflect how partners actually go to market today. Facilities in the US and Sweden are fully networked to deliver a consistent design and testing experience regardless of location, and are designed specifically to help partners accelerate the move from AI and cloud evaluation into deployment and monetization. * Mitel names new channel chief: Mitel has appointed Ben Macdonald as vice president of global channel go-to-market, bringing experience from Owl Labs, Poly, Juniper Networks, and Ekahau. The hire comes as Mitel's own research shows 68 percent of businesses are running communications infrastructure more than seven years old, with 92 percent of modernizing organizations choosing an integrated-hybrid strategy - a dynamic the company says positions its 6,000-plus channel partners at the center of one of the largest communications refresh cycles in a decade. * Cork Cyber wins Pax8 Startup Vendor of the Year: Pax8 recognized Cork Cyber at Beyond26 for its AI-native remediation platform built for MSPs, which remediates threats automatically, reduces ticket volume, and provides financial payback when risks slip through. The award was presented on the Beyond mainstage by Pax8 president Nick Heddy. * Canada's cloud market: A new report from the Canadian Anti-Monopoly Project, covered by CBC News, calls the Canadian cloud computing market "broken," warning that Amazon, Microsoft, and Google control approximately 85 percent of the market. The report argues that even adding domestic sovereign alternatives will not fix the problem without interoperability standards, coining the term "maplewashed dependency" for the risk of trading one lock-in for another. * Pentesting research: New research from Cobalt and Omdia finds that 53 percent of security leaders believe traditional penetration testing is now outdated, with demand growing for continuous, AI-assisted approaches. * iCOUNTER leadership: iCOUNTER has appointed Joel Molinoff, formerly of BlueVoyant and CBS Corporation, as chief operating officer. * DataStrike expansion: DataStrike has expanded its Linux managed services practice by hiring Jon Cain as senior Linux infrastructure engineer to meet growing client demand. Audio Player Read Full Transcript

BlueVoyant has launched BlueVoyant AI, an AI-native Security Operations Center platform designed to prevent, detect and respond to cyber threats autonomously. The platform offers two deployment options: a fully managed service supported by BlueVoyant's SOC team, or a self-service SaaS solution for internal security teams. Built on nearly 10 years of experience across the Microsoft Security ecosystem, BlueVoyant AI automates threat response actions including device isolation, credential revocation and malicious email eradication at machine speed. The platform eliminates alert fatigue by instantly enriching and triaging alerts, reducing false positives and response times. The company has raised $295 million to date and serves over 2,500 customer deployments. BlueVoyant's roadmap includes expanding into identity security, particularly for non-human identities in enterprise environments.