NexusAI, a leading AI research lab, confirmed a significant data breach exposing terabytes of its proprietary training data used for the upcoming Aura-5 large language model. The breach was traced bac
A malicious package impersonating PyTorch Lightning was downloaded over 1,000 times, stealing SSH keys, Kubernetes configs, and other secrets from AI developers. This supply chain attack highlights growing security risks in the AI ecosystem. Are your AI development environments secure?
A major breach at Cognition AI, the company behind the Devin autonomous agent, was disclosed today. Attackers exploited a zero-day vulnerability in a third-party data processing library used in their
In a landmark transatlantic agreement, the United States and the European Union have announced the 'AI Liability Shield' framework, a new set of regulations for developers of high-capability foundatio
A team from Stanford's AI Lab has published a paper detailing a novel jailbreak technique called the 'Cognitive Dissonance' attack. The method involves crafting prompts that create a logical paradox f
Leading AI model provider SynthCorp confirmed a significant data breach that occurred late last month. Attackers exploited a vulnerability in a third-party MLOps platform, gaining access to core infra
AI security firm Metacortex has launched Guardian, a powerful open-source tool designed to help developers secure their large language model (LLM) deployments. Guardian automates the red teaming proce
Emerging AI leader ChronoAI has confirmed a major security breach orchestrated by a sophisticated state-sponsored threat actor. The attack, which occurred over several weeks in January 2025, resulted
The European Commission has officially adopted the AI Safety & Security Act (ASSA), a landmark piece of legislation that imposes stringent security obligations on developers of 'high-risk' AI systems
A paper published by researchers at the Carnegie Mellon AI Safety Institute has introduced a novel jailbreak technique called 'Semantic Reframing'. This method successfully bypasses the safety filters
AI startup Cognition Labs confirmed a critical security breach that resulted in the exfiltration of proprietary model weights for its upcoming 'Nexus-5' generative model, along with a significant port
The U.S. National Institute of Standards and Technology (NIST) has launched ModelGuard, an open-source security framework designed to help organizations evaluate the robustness and security of their A