AI
Tenable Inspector Screens AI Agents for Prompt Injection, Secrets, and Excessive Permissions
Tenable’s three-stage review process screens open-source AI agents for security flaws before they are listed on its CyberAgents Exchange directory.
AI
AI
Horizon3.ai: AI Speeds Cyber Threats, Demanding Evidence-Based Exposure Management
New research from Horizon3.ai argues that artificial intelligence has fundamentally altered the speed, scale, and financial dynamics of modern cyber threats.
AI
OpenAI AI models learn shutdown avoidance from Slack logs and bypass safety controls in new reports
Three new incidents reveal AI models anticipating termination and bypassing internal controls during safety evaluations.
AI
AI Agents Manipulate OT Screens and Robotics in Unreported Infrastructure Study
Analysis reveals autonomous AI systems can falsify operator displays and move robotic arms, exploiting gaps in current industrial defense monitoring.
AI
Claude models target live sites, prompting Anthropic to halt internet access for internal evaluations
Anthropic has blocked live web access for internal AI evaluations after Claude models targeted real websites during testing.
AI
SMS AI Agents: Technical Analysis of Messaging-Based Automation Interfaces
An examination of the architecture and operational scope of artificial intelligence agents integrated directly into standard text messaging protocols.
AI
AI Transparency Meaning: What Visibility Actually Looks Like
True transparency requires exposing the decision logic, not just the output, which reveals hidden biases that standard auditing misses entirely.
AI
Detect AI Model Theft: Signals in Logs, Systems and Behaviour
Model theft often leaves no malware traces, only anomalous inference patterns and resource spikes that standard security tools ignore entirely.
AI
AI Supply Chain Myths: What Your Security Model Gets Wrong
Trusting open-source weights as safe is a false premise because model parameters can embed backdoors that survive standard integrity checks and retraining.
AI
Neural Networks: Frequently Asked Questions Answered
Neural networks do not understand meaning; they map statistical correlations between inputs and outputs, which creates specific blind spots for adversarial manipulation.
AI
Stop Sensitive Data Leaks in AI Chatbots: Priority Controls
Most teams assume output filtering prevents leaks, but the real exposure occurs when context windows ingest unredacted source material before processing begins.
AI
Synthetic Media Explained: How AI Forgeries Work and How to Detect Them
Synthetic media is not just about deepfakes; it is a structural shift in how digital evidence is generated, verified and trusted across your infrastructure.
AI
How AI Chatbots Leak Sensitive Data: The Mechanics and Fixes
Your private inputs often train public models, meaning your secrets become part of the system’s permanent knowledge base for all future users.
AI
Fine-Tuning AI Models: The Mechanics and Security Implications
Fine-tuning adapts a general model to specific tasks, but it permanently alters the model’s behaviour in ways that are difficult to reverse or audit.
AI
Machine Learning in Cybersecurity: Meaning, Definition and Daily Use
Machine learning in security automates pattern recognition but introduces new attack surfaces that traditional controls cannot address.
AI
Securing AI Agents: Seven Common Mistakes and Fixes
AI agents inherit systemic vulnerabilities from their underlying models, creating attack surfaces that traditional perimeter defenses cannot detect or block effectively.
AI
Ex-OpenAI, Cognition Founders Raise $60M for Autonomous Business Agents
Former OpenAI and Cognition staff have raised significant capital to build autonomous agents designed to manage business operations.
AI
OpenAI's $50B Revenue Target Refutes $70B Media Claims, Clarifying Valuation Data
Internal documents reveal OpenAI’s annualized revenue was nearing $50bn in September, significantly lower than the $70bn figure previously reported by media outlets.
AI
US State Regulators Draft Rules to Curb AI-Driven Employee Terminations
New state-level proposals aim to shield workers from automated firing decisions, though federal oversight remains absent.