AzureMicrosoft is urging organizations to stop treating AI agents as isolated pilots and instead manage them as an ongoing operational discipline. The company highlights Microsoft Foundry as the platform for planning, controlling, and optimizing AI agent cost, performance, governance, and ROI at scale.
3 min read · Nov 12, 2026
SecurityMicrosoft has observed multiple threat actors deploying the open source GSocket remote access tool on compromised Linux web servers to maintain access after the original vulnerability is patched. The technique relies on outbound-only encrypted connections through a shared relay network, making traditional IP or domain blocking far less effective and raising the need for host-level investigation and egress monitoring.
3 min read · Nov 11, 2026
SecurityMicrosoft’s FORGE Lab detailed how its MDASH project turned 26,502 Linux kernel scan findings into 59 upstream bug reports and 33 CVEs, with 38 fixes already merged. The research highlights both the promise of AI-assisted vulnerability discovery and the continued need for expert validation, patching, and coordination with upstream maintainers.
3 min read · Nov 11, 2026
SecurityMicrosoft Research says AI models can be backdoored by editing fewer than 10 neurons, creating sleeper agents that pass evaluations but execute attacker commands when triggered. The findings matter because autonomous agents often run with privileged access, making model weight integrity, auditing, and runtime controls critical for enterprise AI security.
3 min read · Nov 11, 2026
SecurityMicrosoft FORGE Lab disclosed several memory safety issues in Rust’s standard library, including cases where safe Rust code could still trigger undefined behavior. The research, aided by the Hyenas AI vulnerability-scanning system, highlights recurring unsafe-code design pitfalls that matter for developers and security teams relying on Rust at scale.
3 min read · Nov 9, 2026
AzureMicrosoft says the next phase of AI depends on infrastructure designed for long-running agentic workloads, not just better models. The company positions Azure as an integrated platform that combines compute, storage, networking, and operations to help organizations run AI agents reliably, cost-effectively, and at scale.
3 min read · Nov 9, 2026
AzureMicrosoft has introduced the Microsoft Copilot harness in Foundry Agent Service as a public preview, giving developers a fully managed way to build agents that can reason, use tools, and complete long-running tasks. The company also announced general availability of long-running resilience for hosted agents, helping teams build more reliable AI workflows without managing the full execution stack themselves.
3 min read · Nov 9, 2026
AzureMicrosoft has introduced Hardware Audit Logging for Azure Cloud HSM, providing hardware-generated audit records for supported HSM management operations. The feature helps security and compliance teams verify administrative activity with HSM-signed evidence, while integrating events into Azure Monitor Log Analytics and Microsoft Sentinel for monitoring and investigations.
2 min read · Nov 9, 2026
Entra IDMicrosoft Entra’s November 2026 updates focus on securing access as users, devices, guests, and AI agents change over time. Key additions include agent and guest lifecycle policies, faster Continuous Access Evaluation based on device and user risk, phishing-resistant MFA improvements, and new governance tools for access drift and application groups.
3 min read · Nov 7, 2026
SecurityMicrosoft Threat Intelligence reports that Storm-3358 is using the ARTEX agentic AI offensive framework to target financial organizations, especially in Southeast Asia. The campaign combines AI-driven reconnaissance and exploitation with RAT deployment, credential theft, and cloud identity abuse, raising the urgency for defenders to strengthen identity, endpoint, and deception-based protections.
3 min read · Nov 7, 2026
Entra IDMicrosoft has outlined its Microsoft Entra presence at Ignite 2026, with breakout sessions focused on securing access for users and AI agents, unified policy management, and faster identity threat response. For IT and security teams, the event offers a practical roadmap for adapting identity and access controls to AI-driven environments.
3 min read · Nov 6, 2026
SecurityMicrosoft reports threat actors are using agentic AI tools to scan public GitHub repositories for exposed secrets and rapidly turn them into cloud compromise. The campaign highlights how leaked application credentials, service principals, and API keys can lead to fast access to Azure, Entra ID, Microsoft 365, and AI resources, making secret rotation and stronger workload identity controls urgent priorities.
3 min read · Nov 6, 2026