A new paper from the Stanford AI Lab details a novel jailbreak technique named 'Recursive Contextual Poisoning' (RCP). Unlike traditional prompt injection methods that rely on single, complex prompts,
AIBW Security DeskCognition AI, a leading provider of enterprise-grade LLM APIs, disclosed a severe security breach that exposed sensitive customer data and, more alarmingly, the proprietary model weights for its flags
AIBW Security DeskResearchers from Stanford's AI Security Lab have published a groundbreaking paper detailing a new jailbreak technique named the 'Cognitive Dissonance Attack.' This method has proven alarmingly effecti
AIBW Security DeskThe AI Security Foundation has announced the official open-source release of the Aegis AI Framework, a comprehensive toolkit designed to help developers secure their LLM-powered applications. Aegis pr
AIBW Security DeskSynthAI, a leading generative AI company, confirmed a critical security breach resulting in the exfiltration of proprietary model weights and millions of customer records. The incident, which occurred
AIBW Security DeskThe U.S. National Institute of Standards and Technology (NIST) has released the final version of its AI Secure Development Framework (AI-SDF), a set of standards and best practices for building secure
AIBW Security DeskAI startup SynthWeave, valued at over $5 billion, has confirmed a significant security breach resulting in the exfiltration of its flagship proprietary text-to-video model weights and over 50 terabyte
AIBW Security DeskA new research paper from Carnegie Mellon University's CyLab has detailed a novel jailbreak technique named 'Cognitive Dissonance'. The attack involves crafting complex, multi-turn prompts that presen
AIBW Security DeskThe U.S. Department of Commerce, through NIST, has issued new binding regulations based on the AI Risk Management Framework 2.0, mandating robust AI watermarking and data provenance reporting for comp
AIBW Security DeskResearchers from Carnegie Mellon University have published a new paper detailing a novel jailbreak technique named 'Contextual Shifting.' The method effectively bypasses state-of-the-art safety alignm
AIBW Security DeskAI video generation startup SynthVerse has disclosed a significant data breach stemming from a misconfigured cloud storage bucket. The exposed dataset contained over 10 million user prompts, associate
AIBW Security DeskThe U.S. Cybersecurity and Infrastructure Security Agency (CISA) has issued a binding operational directive requiring all software vendors providing AI-powered systems to federal agencies for critical
AIBW Security Desk