All news with #microsoft tag

🏆 The Pwn2Own Berlin 2026 contest at OffensiveCon (May 14–16) awarded security researchers $1,298,250 for exploiting 47 zero-day vulnerabilities across browsers, enterprise apps, servers, virtualization, containers, LLMs and local privilege escalation. Competitors earned $523,000 on day one, $385,750 on day two, and $389,500 on day three. DEVCORE topped the leaderboard with $505,000 and 50.5 Master of Pwn points; Cheng‑Da Tsai secured the highest single payout of $200,000 for an Exchange RCE chain.

🔒 A security researcher alleges Microsoft quietly changed Azure Backup for AKS behavior after rejecting his March disclosure and blocking a CVE, arguing the issue required pre-existing administrative access. The reported flaw purportedly allowed a user with only the Backup Contributor role to gain cluster-admin privileges via Trusted Access. Microsoft maintains the behavior was expected and that no product changes were made, yet the researcher observed new permission checks and a shift to manual Trusted Access configuration after disclosure. CERT/CC validated the bug but the CVE process stalled, leaving defenders with limited visibility.

⚠️Microsoft has warned of a zero-day cross-site scripting vulnerability in Exchange Outlook Web Access (OWA) that can be triggered by a specially crafted email. The flaw (CVE-2026-42897) is being actively exploited and affects Exchange Server 2016, 2019, and Server Subscription Edition, while Exchange Online is unaffected. Microsoft has published an automatic mitigation via the Exchange EM Service; administrators should enable EM Service or run the Exchange on-premises Mitigation Tool (EOMT) if servers are air-gapped. The interim mitigations can disrupt OWA features such as calendar printing and inline image display, and a formal security update will be released later.

🔒 Microsoft will change Edge so saved passwords are not loaded into process memory in clear text at startup. Security researcher Tom Jøran Sønstebyseter Rønning disclosed on May 4 that Edge decrypted all stored credentials at launch and released a proof-of-concept showing how attackers with Administrator privileges could dump other users' passwords. Microsoft initially described the behavior as "by design" but now says a defense-in-depth change will roll out across Stable, Beta, Dev, Canary and Extended Stable; the fix is live in Canary and will be in build 148 and newer.

🔐 Google Threat Intelligence Group (GTIG) details UNC6671, operating as "BlackFile," which uses large-scale voice phishing (vishing) and adversary-in-the-middle techniques to bypass MFA and compromise SSO access. The group targets Microsoft 365 and Okta, leveraging Python and PowerShell scripts to automate exfiltration and repurpose valid session cookies to "stream" files. GTIG highlights detection indicators such as python-requests User-Agent mismatches, nonstandard IP infrastructure, and subdomain-based credential-harvesting sites to aid defenders.

⚠️ Microsoft has disclosed a high-severity zero-day, CVE-2026-42897, in on-premises Exchange Server that could allow an attacker to execute arbitrary code by sending a specially crafted email to an Outlook user. The flaw is an XSS vulnerability affecting all supported versions of Exchange 2016, 2019 and Subscription Edition, but not Exchange Online. Microsoft recommends enabling the Exchange Emergency Mitigation (EM) Service, which is applied by default, and provides an alternative manual mitigation via the Exchange On-premises Mitigation Tool for air-gapped environments while patches are developed.

⚠️ Microsoft disclosed a new actively exploited vulnerability, CVE-2026-42897 (CVSS 8.1), a spoofing bug caused by cross-site scripting in on-premises Exchange Server. An attacker can execute arbitrary JavaScript by sending a crafted email that is opened in Outlook Web Access. Microsoft offers a temporary mitigation via the Exchange Emergency Mitigation Service (enabled by default) and provides an EOMT PowerShell script for environments that cannot use the service; Exchange Online is not affected.

🔒 On day one of Pwn2Own Berlin 2026 researchers earned $523,000 exploiting 24 unique zero-days, led by Orange Tsai, who collected $175,000 after chaining four logic flaws to escape the Microsoft Edge sandbox. Windows 11 was rooted three times for new privilege-escalation bugs, and Valentina Palmiotti secured payouts for Red Hat Workstations and an NVIDIA Container Toolkit flaw. The event focuses on enterprise and AI-targeted technologies.

🔔 AWS has added GLM-5.1-FP8 (from Z.ai) and Phi-4-mini-instruct (from Microsoft) to Amazon SageMaker JumpStart, expanding foundation model choices for enterprise workloads. GLM-5.1-FP8 targets agentic software engineering and multi-round optimization for repository-level code, debugging, and long-horizon automation. Phi-4-mini-instruct provides compact, low-latency reasoning across 24 languages and supports function calling for edge and latency-sensitive use cases. Customers can deploy these models via SageMaker Studio or the SageMaker Python SDK in a few clicks.

🛡️ Microsoft Security explains how rising agentic autonomy reorients security from models to how agents are assembled, constrained, and governed inside applications. The post identifies amplified risks—agent hijacking, intent breaking, data leakage, supply chain compromise—and shows why the application layer is decisive because builders fully control permissions, tool access, and failure handling. It recommends concrete design patterns: agents as microservices, least permissions, deterministic human-in-the-loop, and distinct agent identity to limit blast radius and preserve auditability.

🔒 Microsoft Defender research shows AI and agentic applications on cloud-native platforms are frequently deployed with insecure defaults and missing authentication, creating exploitable misconfigurations. Observed exposures include public MCP servers, unsecured Helm chart installs, and unauthenticated agent frameworks that enable remote code execution, credential theft, and access to internal tools. Defender for Cloud can detect exposed Kubernetes services and unsafe deployment patterns to help teams prioritize remediation.

🔒 A researcher known as Chaotic Eclipse (Nightmare-Eclipse on GitHub) published proof-of-concept exploits named YellowKey and GreenPlasma that bypass BitLocker protections and enable local privilege escalation on affected Windows versions. YellowKey abuses the Windows Recovery Environment (WinRE) and NTFS transaction replay to spawn a shell and access encrypted volumes, while GreenPlasma allows arbitrary memory-section creation that can be escalated to SYSTEM. The author said the disclosures were driven by dissatisfaction with Microsoft's handling of reports. Microsoft says it investigates and supports coordinated disclosure.

🔧Microsoft outlines its sustained investment in PostgreSQL through upstream contributions, managed services, developer tools, and community programs. The post highlights 345 commits to the latest PostgreSQL release, active Microsoft committers working upstream, and service offerings such as Azure Database for PostgreSQL and Azure HorizonDB. It also emphasizes AI integrations like vector search and model invocation alongside IDE tooling and community engagement.

🔧 Microsoft released a cumulative update addressing a BitLocker recovery issue that caused some systems to prompt for recovery keys after installing the April 2026 security updates. The KB5089549 patch fixes the problem on Windows 11 25H2 by correcting boot-file update behavior tied to certain TPM validation and invalid PCR7 settings. Administrators are advised to remove the Configure TPM platform validation profile for native UEFI firmware configurations Group Policy before broad deployment and to confirm BitLocker bindings use the PCR7 profile.

🔧 Microsoft has applied a service-side fix for a Windows Autopatch bug that caused driver updates restricted by administrative policies to be deployed on some EU-managed Windows devices. The issue affected a limited set of client platforms, including Windows 11 25H2, 24H2, and 23H2. Impacted systems experienced unexpected reboots and, in some cases, system failures depending on the installed drivers. Microsoft says no client-side action is required.

🛡️ Microsoft introduced MDASH (multi-model agentic scanning harness), a model-agnostic AI system in limited private preview designed to discover, validate, and prove exploitable defects in large codebases. The system orchestrates more than 100 specialized agents across frontier and distilled models in a structured pipeline that builds threat models, runs auditor and debater stages, groups equivalent findings, and proves vulnerabilities. Microsoft reports MDASH uncovered 16 issues fixed in this month’s Patch Tuesday, including two critical Windows networking and authentication flaws.

🔍 Microsoft disclosed MDASH, an AI-driven vulnerability discovery system that found 16 previously unknown Windows flaws, including four critical remote code execution bugs that were patched as part of the May 12 Patch Tuesday release. Built by the Autonomous Code Security and Windows Attack Research teams, the platform orchestrates more than 100 specialized AI agents across multiple models to scan, validate and construct triggering inputs before human review. Microsoft said MDASH is intentionally model-agnostic and will enter private enterprise preview next month.

⚠️ Microsoft confirmed that a recent service update introduced a configuration change preventing some customers from downloading and installing Office on Windows 365 Cloud PCs. The issue, tracked as WP1309017 and first acknowledged on May 12, is being investigated and a fix is in development. Microsoft said validation and deployment of the fix will take time and expects a further update on Friday. Affected users can manually download Office from the Microsoft 365 page while remediation proceeds.

🔒 Microsoft released patches for 138 vulnerabilities across its product portfolio, including 30 Critical and 104 Important flaws, with none currently listed as publicly known or under active attack. The update spans privilege escalation, remote code execution, information disclosure, and spoofing issues, and includes a recently patched AMD CPU isolation flaw (CVE-2025-54518). Notable high-risk fixes include CVE-2026-41096 (Windows DNS heap overflow) and several Critical issues in Azure, Dynamics 365, Hyper-V, and Office. Administrators are urged to prioritize updates, rotate Secure Boot certificates before the June 26, 2026 deadline, and follow mitigation guidance such as reducing internet exposure and enforcing MFA.

🔒 Microsoft released its May Patch Tuesday fixing 120 CVEs, including 17 critical flaws. The update addresses 14 RCEs, two elevation of privilege bugs and one information disclosure issue, with the majority of fixes covering EoP and RCE types. Microsoft credited its WARP team and an agentic AI system, MDASH, with discovering 16 of the issues. Administrators are urged to prioritize high-risk fixes such as CVE-2026-41089.