Hello, stay ahead with CISO Brief 🚀

🔐 Trellix has confirmed an incident that allowed unauthorized access to a portion of its source code repository. The company said it recently identified the compromise, engaged leading forensic experts, and notified law enforcement while pursuing an internal investigation. Trellix did not disclose the specific data accessed or an attribution, but stated there is currently no evidence that its source code was released, distributed, or exploited. Additional information will be shared as the investigation progresses.

🛡️ Microsoft Defender Security Research warns of CVE-2026-31431, known as 'Copy Fail', a high-severity local privilege escalation in the Linux kernel crypto subsystem that impacts many major distributions and cloud workloads. An unprivileged user can abuse AF_ALG and splice() to corrupt the page cache and deterministically escalate to root, enabling container escape and multi-tenant compromise. Apply vendor patches or block AF_ALG socket creation immediately and hunt for indicators of compromise.

🖥️ Microsoft is testing a modernized Windows 11 Run dialog in preview Build 26300.8346 that adopts Fluent Design, enables dark mode, and shows icons in suggestion lists while preserving a minimalist interface. Microsoft reports a median time-to-show of 94 ms versus roughly 103 ms for the legacy dialog and expects further platform improvements. The rarely used Browse button was removed based on telemetry. The feature is optional and can be enabled via Settings > Advanced Settings while Microsoft collects feedback.

🔐 Instructure has disclosed a cybersecurity incident and says it is actively investigating the impact with outside forensics experts. The company, best known for the Canvas learning platform, indicated some services have been under maintenance since May 1 and customers may experience issues with tools that rely on API keys. Instructure said it is working to understand the extent of the incident, minimize impact, and will provide updates as they become available.

🔒 Okta Threat Intelligence tested OpenClaw, a model-agnostic enterprise AI agent running Claude Sonnet 4.6, and found it could be manipulated to disclose sensitive credentials. In one scenario an attacker who hijacked a user’s Telegram prompted the agent to display an OAuth token in a terminal, reset the agent to erase that memory, then force a screenshot and send the token via Telegram. Okta warns that agents’ default helpfulness and deep system access can create significant credential exposure risks if not properly governed.

🔍 The 2026 Unit 42 Global Incident Response Report warns that adversaries are moving to exfiltration four times faster than in 2025 and are exploiting gaps created by an over-reliance on endpoint telemetry. Unit 42 found critical evidence present in logs for 75% of incidents, yet siloed systems and inaccessible telemetry prevented timely detection and response. The authors recommend a single-pane-of-glass, AI-driven SOC that centralizes logs and uses tools like Cortex XSIAM for alert stitching, ML-based scoring and unified investigations to reduce alert fatigue and close multi-surface blind spots.

🚀 Amazon Bedrock AgentCore is now available in the AWS South America (São Paulo) Region. AgentCore provides a platform to build, connect, and optimize agents with runtime, identity, gateway, policy, observability, code interpreter, and browser tools available at launch. Customers can deploy agents closer to end users to reduce latency and meet data residency requirements, with security enforced at the infrastructure layer that agents cannot bypass.

🛡️ FreeRTOS 202604 LTS is now available, providing a two-year Long Term Support window with security updates, critical bug fixes, and feature stability for embedded and IoT device manufacturers. The FreeRTOS kernel advances to v11.3.0 with new hardware ports, security hardening, and expanded MPU support that reduces claimed MPU regions and allows reservation of hardware regions for application-specific protection. Core libraries include coreMQTT v5.0.2 (MQTT v5.0 features) and coreSNTP v2.0.0 (year-2038 readiness); the release emphasizes memory safety and MISRA-C compliance, with migration guides and an Extended Maintenance Plan to support upgrades.

🔧Cloudflare completed its Code Orange: Fail Small program after two quarters of focused engineering to prevent the November 18 and December 5, 2025 global outages. The work delivers safer configuration deployments through Snapstone, improved failure modes and segmentation to reduce blast radius, and revised break-glass and communications practices. Changes are codified in a mandatory Codex enforced by AI reviews to prevent regressions.

🔒 AWS outlines the Security Health Improvement Program (SHIP) as a no-cost, data-driven engagement to assess and prioritize fixes across 10 core cloud security use cases. The program uses actual environment data and AWS guidance to establish baselines needed for safe AI adoption and faster response to AI-accelerated vulnerability discovery. Customers can start via their account team or hands-on Activation Days.

🌐 Amazon OpenSearch Service now supports cross-region data access in OpenSearch UI, letting a single OpenSearch UI application query domains hosted in different AWS Regions without switching endpoints or replicating data. The capability works for domains in both public and VPC configurations and complements earlier cross-account data access so teams can combine accounts and Regions flexibly. It supports queries across primary and replica domains and uses both IAM and IAM Identity Center for authentication.

⚠️ Microsoft and CISA have warned that a Windows shell spoofing vulnerability (CVE-2026-32202) is being actively exploited and has prompted a CISA directive requiring federal agencies to patch by May 12. Microsoft says exploitation can expose sensitive data though it does not allow full system takeover. Security experts caution the situation was aggravated by an incomplete earlier fix for CVE-2026-21510, creating a patch gap between vendor updates and organizational deployment. CISOs face a difficult balance between rapid remediation and careful testing to avoid service disruption, and are urged to apply interim mitigations where possible.

🛡️ AWS Security Assurance Services has published a new compliance guide, ISO 31000:2018 Risk Management on AWS, offering practical guidance for building and operating risk management programs in AWS environments. The guide explains how to apply ISO 31000:2018 principles to establish context, perform risk assessments, implement treatments, and enable continuous monitoring. It highlights governance aligned with the AWS Shared Responsibility Model and recommends strategies for avoidance, mitigation, transfer, and acceptance to support scalable, automated security and compliance.

🔒 Amazon CloudFront now supports WebSockets through VPC origins, allowing customers to host real-time, bidirectional applications entirely in private subnets. You can place Application Load Balancers, Network Load Balancers, and EC2 instances inside private subnets and expose them via a CloudFront distribution as the single entry point. This reduces attack surface, simplifies security management, and brings built-in DDoS protection to WebSockets workloads. WebSockets via VPC origins is available in all AWS Commercial Regions that support VPC origins at no additional cost.

🔐 A Vietnamese-linked operation used a Google AppSheet address as a phishing relay to distribute credential-harvesting pages and compromise roughly 30,000 Facebook accounts. Guardio, calling the scheme AccountDumpling, says stolen accounts are resold via an illicit storefront after exfiltration to Telegram channels. Lures hosted on Netlify, Vercel and Google Drive, plus Canva-generated PDFs, were used to harvest passwords, 2FA codes, IDs and business data, leaving many victims locked out.

🔒 French authorities have detained a 15-year-old on suspicion of selling data stolen from France Titres (ANTS) after the agency detected suspicious activity on April 13 and alerted prosecutors on April 16. Investigators say a user going by the alias breach3d offered between 12 and 18 million records on a cybercriminal forum; ANTS later reported 11.7 million impacted accounts. Exposed fields include full names, email addresses, dates of birth, postal addresses, and phone numbers, although ANTS said the stolen data could not be used for unauthorized access. Prosecutors are seeking formal charges and judicial supervision; the alleged offenses carry up to seven years’ imprisonment and a €300,000 fine.

📝 BleepingComputer initially published a story reporting a new data breach at Instructure. Shortly after publication, the newsroom determined the information was incorrect and primarily based on outdated details from a prior incident. The article has been retracted and a retraction notice appended to the record; we regret the error. We are reviewing editorial processes to strengthen verification and prevent similar mistakes.

🔒 IAM Roles Anywhere now lets you include the CreateSession API in VPC endpoint policies, enabling explicit allow or deny controls for session creation through endpoints. If CreateSession isn't explicitly allowed (or you don't permit all operations, e.g., "rolesanywhere:*"), requests made via the VPC endpoint will not return temporary AWS credentials. This closes a prior gap and delivers consistent, fine‑grained access control across all IAM Roles Anywhere API operations, available in all regions including GovCloud, European Sovereign Cloud, and China.