Multi-Model AI Is Creating a Routing Headache for Enterprises
Enterprises are treating AI inference as a core production workload, with 78% operating their own services and averaging seven models per deployment, according to F5’s 2026 State of Application Strategy Report. Multi‑model inference spreads across hybrid multicloud environments, demanding new traffic‑management, observability, and security controls. Control planes now dictate where inference requests are routed, how they are protected, and how costs are managed. The shift creates routing headaches but also opportunities for unified delivery and governance solutions.
A Study of 1,000 Android Apps Finds a Privacy Policy Logging Gap
A new study of 1,000 Android apps reveals a stark gap between privacy policies and actual logging practices. Fewer than one‑third of policies mention logging, and only four policies accurately disclose the sensitive data found in runtime logs. The research...
GDPR Works, but only Where Someone Enforces It
A cross‑country measurement study of 525 popular websites shows that GDPR‑style opt‑in rules dramatically lower web tracking, but only where regulators actively enforce them. Users in Germany and Spain experienced roughly 50% fewer tracker connections than visitors from non‑EU regions,...
What the EU AI Act Requires for AI Agent Logging
The EU AI Act classifies AI agents that make credit, hiring, health or emergency decisions as high‑risk, triggering strict logging obligations. Articles 12‑13 require automatic, tamper‑evident logs covering risk events, post‑market monitoring and operational data for the system’s entire lifecycle, with...
Review: The Psychology of Information Security
Leron Zinatullin’s second‑edition book argues that security controls crumble when they ignore the people who must use them. Drawing on organizational psychology, change management and usability research, the text maps risk management fundamentals to practical policy design and behavioral change....
Bringing Governance and Visibility to Machine and AI Identities
AppViewX’s CEO Archit Lohokare says the rapid rise of enterprise AI has merged machine and AI‑agent identities into a single, exponentially growing security problem. To address this, AppViewX has partnered with Eos to layer agentic governance on top of its...
What to Do in the First 24 Hours of a Breach
Help Net Security released a video featuring CYGNVS CEO Arvind Parthasarathi outlining a ten‑step framework for handling a cyber breach. The first five steps focus on preparation, including establishing an out‑of‑band communication channel, mapping internal stakeholders, engaging external legal and...
Does Anthropic Deserve the Trust of the Cybersecurity Community?
Anthropic positioned itself as the trustworthy AI alternative to OpenAI, publishing a Responsible Scaling Policy and launching Claude Code Security in early 2026. In February, the company released RSP 3.0, replacing absolute safety guarantees with relative commitments tied to competitor behavior....
Airbus CSO on Supply Chain Blind Spots, Space Threats, and the Limits of AI Red-Teaming
Airbus Chief Security Officer Pascal Andrei warns that the aerospace and defense supply chain’s deepest vulnerabilities now reside in sub‑tier suppliers and the digital threads linking them. He highlights Airbus’s shift toward a collaborative, intelligence‑led model, tighter integration of security...