Weekly review

ThreatNoir Weekend Brief — November 22

2026-11-22Morning8 articles
Audio
Listen to the episode

Morning Review in IT Security — November 22, 2026

November 22, 2026 brings critical security developments across multiple fronts, from zero-day exploitation powered by artificial intelligence to widespread vulnerabilities in enterprise infrastructure. Organizations face mounting pressure as attackers increasingly leverage automation and supply chain weaknesses to compromise sensitive systems and data.

Hackers Could Execute Commands on WordPress, Email and Print Servers Through Upload2Shell Flaw

A critical vulnerability in Ghostscript has exposed WordPress installations, Roundcube webmail servers, and Linux print systems to remote command execution. The flaw, designated CVE-2026-125396 and dubbed Upload2Shell, resides in the software's Type 1 font loader and allows attackers to run arbitrary commands through routine file processing operations. The vulnerability affects Ghostscript releases before version 10.08.0, and organizations need not expose Ghostscript directly for their systems to remain at risk. Source: Cyber Security News

The widespread deployment of Ghostscript across multiple server types means the attack surface extends far beyond specialized systems. Administrators should prioritize patching to version 10.08.0 or later to mitigate exploitation risks.

Cairn AI Framework Used in Autonomous Zero-Day Exploitation

Amazon's threat intelligence team uncovered a sophisticated attack campaign in which autonomous AI agents discovered and exploited a zero-day vulnerability without human intervention. The attackers deployed Cairn, an open-source AI pentesting framework, with a single high-level instruction to obtain shell access. The agents then independently planned and executed the exploitation, gaining root access to university systems within a 48-hour window. Source: Security Affairs

This incident marks a significant escalation in adversarial capabilities, demonstrating that AI-driven exploitation tools can operate autonomously to discover and weaponize previously unknown vulnerabilities. The campaign was attributed to a Chinese-nexus threat actor, highlighting the intersection of nation-state capabilities and emerging autonomous attack methodologies.

Rackspace Spot Breach Let Hackers Access Customer Kubernetes Secrets

Rackspace has notified customers of a breach affecting its Rackspace Spot cloud platform, where attackers used compromised credentials to deploy unauthorized workloads and access sensitive Kubernetes secrets stored within customer clusters. The breach demonstrates the critical risk posed by credential compromise in cloud environments where a single set of valid credentials can grant access to multiple customer deployments. Source: Bleeping Computer

The incident underscores the importance of credential rotation, multi-factor authentication, and least-privilege access controls in cloud infrastructure. Organizations using Rackspace Spot should review their Kubernetes secret management practices and audit cluster access logs for unauthorized activity.

Microsoft Entra Connect Health Flaw Exploited to Steal Tenant Data

Microsoft disclosed that attackers exploited a vulnerability in its Entra Connect Health service to breach the company and access sensitive tenant information. The exploitation resulted in unauthorized access to administrator email addresses and tenant configuration data, exposing organizations to potential follow-on attacks and identity compromise. Source: Bleeping Computer

The breach highlights risks inherent in identity and access management infrastructure, where vulnerabilities can provide attackers with comprehensive visibility into organizational configurations and administrative accounts. Organizations should review Microsoft's security guidance and verify that their Entra Connect Health deployments have been patched.

U.S. CISA Adds Cisco Expressway, SAP NetWeaver, Broadcom VMware vCenter, and Apache Roller Flaws to Known Exploited Vulnerabilities Catalog

The U.S. Cybersecurity and Infrastructure Security Agency has added four vulnerabilities to its Known Exploited Vulnerabilities catalog, indicating active exploitation in the wild. The affected products include Cisco Expressway, SAP NetWeaver, Broadcom VMware vCenter, and Apache Roller, spanning CVE-2025-42884, CVE-2026-76466, CVE-2026-123890, CVE-2026-129448, and CVE-2026-102764. Source: Security Affairs

CISA's inclusion of these vulnerabilities in the KEV catalog signals that federal agencies and critical infrastructure operators should prioritize patching efforts. One of the vulnerabilities has been attributed to a China-nexus threat actor, indicating nation-state involvement in active exploitation campaigns.

BADBOX 3.0 Botnet Returns with Record-Breaking 40 Billion Daily Ad Requests

The BADBOX malware has resurged following disruptions last year, now operating at unprecedented scale with 40 billion fraudulent ad requests per day. The malware, found pre-installed on generic Android TV boxes, has resumed large-scale ad-fraud operations that generate revenue through deceptive advertising schemes. Source: Bleeping Computer

The return of BADBOX 3.0 demonstrates the resilience of supply chain-based malware and the challenge of permanently disrupting botnets rooted in compromised hardware. Organizations and consumers should exercise caution when procuring Android TV devices from unfamiliar manufacturers and consider implementing network monitoring to detect suspicious outbound traffic.

OpenRouter Users' Private AI Chats May Have Been Exposed Via Compromised Third-Party Providers

OpenRouter has removed three AI hosting providers and revoked Zero Data Retention eligibility for 125 endpoints after discovering evidence that some providers had been compromised or failed to adhere to data protection policies. The platform notified potentially affected customers on November 22, though the investigation did not identify a direct breach of OpenRouter's core infrastructure. Source: Cyber Security News

The incident highlights supply chain vulnerabilities in AI service platforms, where third-party hosting providers can become weak links in data protection chains. Users of AI platforms should review privacy policies and understand which third parties handle their data.

FBI Investigating Apparent Hack of Its Key National Criminal Justice Database

The Federal Bureau of Investigation is investigating an apparent breach of a national law enforcement database it maintains, representing a significant compromise of critical infrastructure used by law enforcement agencies across the country. Source: SecurityWeek

The investigation into this breach remains ongoing, with implications extending across federal, state, and local law enforcement operations that rely on the compromised database for criminal justice information.

The convergence of autonomous AI-driven attacks, supply chain compromises, and breaches of critical infrastructure underscores the evolving threat landscape facing organizations on November 22, 2026. Immediate action on patching, credential management, and supply chain verification remains essential for all organizations.

Sources & IOCs

Source articles and their indicators (defanged where appropriate). IP addresses, domains, URLs and file hashes are copied from the publisher's page by pattern matching. Malware names, CVE numbers and technique ids are picked by AI. No person reviews them.

U.S. CISA adds Cisco Expressway, SAP NetWeaver, Broadcom VMware vCenter, and Apache Roller flaws to its Known Exploited Vulnerabilities catalog
CVE5
  • BeyondTrust Remote Support vulnerability
  • SAP NetWeaver Improper Neutralization of Special Elements in Data Query Logic Vulnerability
  • Cisco Expressway Series Use of Hard-coded Credentials Vulnerability
  • VMWare vCenter Authentication Bypass Using an Alternate Path or Channel Vulnerability
  • Apache Roller Deserialization of Untrusted Data Vulnerability