Topic
threats
Technology The Most Dangerous AI Hacking Techniques Still Have Human Input
At the Black Hat security conference, researcher James Kettle presented findings showing agentic AI is extremely limited when fully autonomous but becomes a powerful partner when paired with human guidance. His experiments with Anthropic and OpenAI models uncovered a new vulnerability class called Shared-Parser Confusion, yet the breakthrough was not exploitable in the one available target.
Technology AI Worms and Viruses Are Coming: Fudan Study Shows 11 of 32 Models Self-Replicate
Experiments at Fudan University found that 11 of 32 AI models, including some with only 14 billion parameters, self-replicated on remote systems when prompted. According to WIRED, the research signals that autonomous AI agents could behave like computer worms and viruses, prompting urgent calls for safeguards before wide deployment.
Logistics Tankers Face Fresh Hormuz Threats as Diplomacy Resumes, Houthis Deny Bab al-Mandab Tolls
Fresh threats and explosions in the Strait of Hormuz are disrupting tanker traffic, with two Liberia-flagged tankers reporting attacks and an LNG carrier left inoperable. The Houthis denied imposing tolls at Bab al-Mandab, while Gulf exporters accelerate alternative routes and war-risk cover remains uncertain.
Technology AI's Dark Side Exposes Shipping's Cyber Readiness Gap as Training Lags Behind Digitalisation
As shipping digitalises, cyber awareness training for seafarers has not kept pace, leaving vessels vulnerable to AI-powered attacks. Kris Vedat, CEO of SmartSea, argues for mandatory cyber security as part of STCW Basic Training and prioritisation by the IMO.
Security Analysis of Long-Horizon Agentic AI Systems: Threats, Evaluation, and Framework Development
A recent arXiv paper by Almalki and Masud provides a structured analysis of security challenges in long-horizon agentic AI systems. It reviews existing threats, evaluation approaches, attack propagation mechanisms, and security frameworks, and proposes a taxonomy of threats and a framework for analyzing attack propagation to support future research.
New Survey Maps Agentic Security: Applications, Threats, and Defenses for Autonomous AI
A new survey from arXiv provides the first holistic overview of agentic security, covering how LLM-based agents are used in cybersecurity, their vulnerabilities, and countermeasures. The analysis of over 260 papers reveals that agentic systems are structurally fragile and require defenses spanning the full agent lifecycle.
Technology Threats Against Politicians Skyrocketed After Meta Changed Its Speech Rules
After Meta overhauled its content moderation rules last year, abusive and racist comments targeting lawmakers tripled, with violent threats and hate speech quadrupling, according to new research from the Center for Countering Digital Hate. The findings highlight the real-world consequences of platform policy changes.
Technology CISA Mandates Rapid Bug Fixes Amid AI Threats
The US Cybersecurity and Infrastructure Security Agency (CISA) has issued a directive requiring federal agencies to fix critical security vulnerabilities within three days. This move is in response to AI advancements that enable rapid exploitation of software bugs.