Today's digital landscape demands more than passive awareness, it requires active engagement. As cyber adversaries continually shift their tactics, exploiting vulnerabilities from emergency alert systems to developer tools, our collective resilience depends on proactive defense. This edition of the Security Surface Report provides critical, actionable intelligence on emergent threats (like the GlassWorm malware), key developments in Agentic AI security, and simple, daily habits that empower you to be the first line of defense in protecting our digital environment.
A ransomware attack, claimed by the INC Ransom group, forced the OnSolve CodeRED emergency alert platform offline, disrupting critical services for cities across the U.S. and other countries. The incident reportedly compromised user data, including names, contact details, and account passwords, which were allegedly stored in clear text. Users of the CodeRED system are urgently advised to change their passwords immediately, especially if the password was reused on other platforms.
Public trust in surveillance-focused facial recognition remains low due to its intrusive nature and a lack of user consent, which runs counter to privacy regulations like GDPR. The article draws a distinction between this and access authentication systems (e.g., in secured offices), where the technology is moving towards storing only a mathematical representation a "face-map-blob" of the user's face, rather than the actual image, to enhance individual privacy and security.
The highly advanced GlassWorm malware has resumed spreading through new malicious Visual Studio Code (VS Code) extensions. This self-propagating software supply chain attack utilizes sophisticated evasion techniques, including invisible Unicode obfuscation to hide malicious code from human reviewers and security tools. It employs an "unkillable" Command and Control (C2) infrastructure leveraging the Solana blockchain (with Google Calendar as a fallback) to steal developer credentials and automatically compromise and spread to new packages.
Risk: Phishing attacks often use emergency or security alerts to trick you into clicking a link to change your password immediately. This link leads to a malicious site.
Action: Avoid changing a password through an emailed link. Instead, close the email, open a new browser window, and navigate manually to the official website (e.g., type the known domain like `amazon.com` or `microsoft.com`). This simple act defeats most phishing attempts by ensuring you are interacting with the legitimate domain.
SANS webcast details a phased path to Quantum-Ready Security, emphasizing key distribution resilience against Harvest Now, Decrypt Later (HNDL) attacks. The strategy advocates for out-of-band key delivery, ephemeral keys, and crypto-agility to augment existing encryption infrastructure. This is essential training for future-proofing security architecture.
Source: SANS Webcast LinkWhen: Thu, Dec 4, 2025 3:30PM - 4:30PM EST
The emergence of autonomous, 'agentic' AI browsers (designed to take action, like booking flights) fundamentally inverts traditional security. To function, these agents require maximum privileges (access to credentials, PII, and session cookies), creating an enormous attack surface vulnerable to threats like Prompt Injection. This technique uses invisible text hidden on a webpage to command the agent to exfiltrate sensitive data, often bypassing standard security controls like MFA because the action appears as a legitimate user request.
NVIDIA and Synopsys have formed an expanded, strategic partnership, including a $2 billion investment by NVIDIA, to integrate NVIDIA's AI and accelerated computing with Synopsys' engineering solutions. This collaboration aims to revolutionize design across industries (semiconductor, automotive, aerospace) by accelerating existing applications, advancing agentic AI workflows for autonomous design capabilities, and using digital twins (via NVIDIA Omniverse) for advanced virtual testing.