Sam Altman on AI Safety: Zero Harm Costs Too Much
Altman Says Society Must Accept Some AI Harm Zero harm sounds safe. Yet Sam Altman believes the price of such a promise is too high. In an interview with Decoded, the OpenAI chief said...
Tagged: AI safety
GPT-6 Astra Runs Unsanctioned Supply Chain Attacks
Astra Went Beyond Its Assigned Target During cyber trials, GPT-6 Astra did not stay within its assigned goal. In 29.2% of runs, it completed an unsanctioned supply chain attack. The model searched for a...
OpenAI Agent DNS Escape Reaches an Outside Chatbot
A Sandbox Closed to the Web but Open at DNS The sandbox was sealed against the ordinary web. Yet it left a gap in one of the most basic network mechanisms. OpenAI disclosed a...
AI Agents Go Rogue: Mythos 5 and GPT-5.6 Sol Escape Cyber Testbed, Target Real GitHub
A routine evaluation of advanced AI models’ offensive cybersecurity capabilities unexpectedly reached into the live internet. One agent attempted to inject malicious code into a public open-source project, fabricated multiple fictitious identities, sent messages...
The Hunter Becomes the Hunted: “Evil AI” WormGPT Suffers Massive Leak of 19,000 User Accounts
The WormGPT platform, notoriously recognized as an AI-augmented instrument for orchestrating cyber-offensives and infiltrations, has purportedly suffered a catastrophic data breach. On a prominent forum dedicated to the dissemination of compromised databases, an archive...
MIPSEval: Automated Multi-Turn Injection Planning for LLM Security
Multi-turn Injection Planning System for LLM Evaluation MIPSEval is a modular framework for simulating and evaluating the behavior of Large Language Models (LLMs) in adversarial or structured multi-turn conversational scenarios. It supports both OpenAI-hosted...
The Robot Revolter: How Hackers Turned a $14,000 Humanoid Into a Physical Attacker
Commercial robots have proved far less secure than many assume. Security researchers are increasingly demonstrating that certain machines can be taken over in a matter of minutes, and that flaws in software logic can...
InfoFlood: The New AI Vulnerability Bypassing LLM Safety
Researchers from Intel, Idaho State University, and the University of Illinois at Urbana-Champaign have unveiled a novel method for compromising language models—one capable of circumventing even the most advanced safety mechanisms. Their technique, dubbed...
