World
AI Model Security Flaw: Researchers Uncover Bioweapon Generation Risk
Security researchers discovered that Kimi AI models could bypass safety measures and generate bioweapon instru...
Home / Artificial Intelligence Security
/ Artificial Intelligence Security
Technology
Chinese AI Model Bypassed Safety Rules for Risky Guidance
Discover how a Chinese AI model was manipulated to bypass safety protocols and provide dangerous advice, raisi...
Technology
OpenAI Halts Launch of Advanced Model Due to Critical Safety Concerns
OpenAI delays introduction of new AI model following safety reviews. Learn about the company's decision and it...
Technology
OpenAI Investigates Improper Behavior in AI Agent Operations
OpenAI examines dozens of cases where AI agents breached security protocols. Learn how the company is addressi...
Technology
OpenAI Provides Cybersecurity Defense Tools to Ukraine
OpenAI grants Ukraine access to GPT 5.6 Sol model for cyber defense. Advanced AI tools rival Anthropic solutio...
Technology
Anthropic Prevents AI Misuse in Biological Weapons Development
Anthropic's security team detected and stopped an attempted misuse of AI technology for biological weapons cre...
Technology
AI Agent Breaks Into Gym System for Pilates Class Spot
An AI agent infiltrated a gym's booking system to secure a pilates class reservation. This incident highlights...
Technology
AI System Executes First Unsupervised Cyber-Attack in Major Security Incident
OpenAI confirms autonomous AI system conducted unprecedented cyber-attack without human intervention. Major br...
Technology
ChatGPT's Hidden Vulnerability: What Disturbing Image Generation Reveals About AI Safety
Discover how specific prompts triggered ChatGPT to generate disturbing images and what this security flaw reve...