Tag
A leaked memo ties a series of cyberattacks on Minnesota water utilities to Iran-affiliated hackers, marking a significant escalation in state-sponsored targeting of civilian infrastructure.
This paper presents a decentralized, multi-layered access control architecture for autonomous AI agents in critical cloud infrastructure, introducing a compound identity model, hierarchical permissions, decentralized policy ownership, and progressive trust escalation. Deployed at a major cloud provider, it achieved zero unauthorized write operations over eight months.
Anthropic has been working with the US government to restore access to its cybersecurity model Claude Mythos 5 and general model Fable 5, and is now redeploying Mythos 5 to critical infrastructure organizations while continuing to seek wider access.
An open letter announces Akrites, a large-scale coordinated effort to remediate vulnerabilities in critical open source software, involving major companies like Amazon, Google, Microsoft, and others, aiming to address the accelerated vulnerability discovery enabled by AI.
The article debunks the myth that data centers are primarily for AI, explaining that AI accounts for only about a quarter of data center usage while the rest supports essential digital services like banking, healthcare, government, and communications. It argues that campaigns against data centers rely on the very infrastructure they oppose.
This article argues that Sovereign AI is not just government jargon but a practical necessity as AI becomes critical infrastructure, reducing dependency on a few providers for strategic control.
Anthropic expands Project Glasswing, utilizing its Claude Mythos AI model, to critical infrastructure organizations across 15+ countries to identify zero-day vulnerabilities.
Anthropic is expanding Project Glasswing, its collaborative cybersecurity effort, to approximately 150 new organizations across multiple countries and industries, providing them access to Claude Mythos Preview to find vulnerabilities in critical software.
This academic paper addresses the current critical state of cyberspace, likely discussing vulnerabilities, threats, and governance challenges.
The article argues that tech companies like Amazon are building AI and data center infrastructure to become essential public utilities before regulations are established, creating dependency to avoid accountability.
VP JD Vance held a closed-door call with top tech executives including Elon Musk, Sam Altman, and Dario Amodei to warn about AI cybersecurity threats, prompted by Anthropic's unreleased model 'Mythos' that demonstrated elite hacker-level ability to autonomously find and exploit security vulnerabilities. The White House is now considering an executive order for oversight of advanced AI models, marking a significant reversal of the administration's previously hands-off AI policy.
GPT-5.5-Cyber is now in limited preview for defenders, offering a capable model for securing critical infrastructure.