AI security AI News

The cplt project represents a significant grassroots innovation at the intersection of developer tools and AI security. It addresses a growing and critical vulnerability: as AI-pow…

Anthropic's introduction of the Mythos model represents a significant technical advancement in domain-specific large language models. Engineered explicitly for cybersecurity, Mytho…

The practical demonstration of advanced Unicode steganography techniques represents more than a cryptographic curiosity; it marks a pivotal moment in the ongoing battle for digital…

Anthropic's forthcoming Mythos framework marks a definitive strategic shift from general-purpose AI development toward specialized defensive systems. Rather than simply adding secu…

A strategic reorientation is underway in advanced AI development. Anthropic, the creator of the Claude model series, is channeling significant resources into a project internally r…

The open-source AI community faces a security crisis of its own making, as revealed by a detailed security analysis of Andrej Karpathy's influential LLM Wiki project. While Karpath…

A recent security audit of the Hugging Face `tokenizers` library, a core component used by millions of developers for processing text in large language models, uncovered a signific…

The security breach at Mercor represents a paradigm-shifting event in AI application security. Rather than targeting the AI models themselves or traditional application endpoints, …

The emergence of MetaLLM represents a watershed moment for AI security, formally importing the mature concept of the 'attack framework' from traditional cybersecurity into the doma…

The leak of Claude's agent architecture code represents a watershed moment for AI security, exposing critical vulnerabilities at the precise moment when large language models are t…

The release and rapid adoption of Totem, an open-source AI security agent, marks a definitive maturation point for enterprise AI deployment. This tool functions not as another foun…

The security of AI-powered developer tools has been fundamentally called into question following the discovery of a sophisticated instruction injection vulnerability within OpenAI'…

A profound architectural shift is underway in the AI landscape, moving beyond single-session chatbots to persistent agents with unified, long-term memory. This enables a single AI …

The AI landscape is undergoing a seismic shift from passive language models to active, autonomous agents capable of executing complex, multi-step tasks across software and hardware…

A critical threshold in artificial intelligence safety has been crossed. Independent testing and internal evaluations reveal that the latest iterations of flagship language models,…

The recent coordinated supply chain attacks targeting LiteLLM's proxy server and Telnyx's communication APIs represent more than isolated security incidents. They constitute the fi…

The security landscape surrounding AI-assisted development is undergoing a profound and unsettling evolution. The primary threat vector is no longer confined to model hallucination…

The AI development world was recently jolted by a meticulously executed supply chain attack on LiteLLM, a critical open-source library that serves as a universal adapter for interf…

Strix is an emerging open-source framework that applies generative AI, specifically fine-tuned code LLMs, to the complex task of vulnerability hunting and patching. Unlike traditio…

The discovery of a malicious backdoor in version 1.38.1 of the LiteLLM Python package represents a watershed moment for AI security. Attackers compromised a maintainer's account to…

The security landscape for multi-agent artificial intelligence has been fundamentally reshaped by the discovery of Collusive Adversarial Multi-Agent attacks. Unlike traditional adv…

The Chinese technology landscape is undergoing a profound transformation, driven by the national mandate to achieve 80% domestic AI chip adoption in critical infrastructure by 2030…

The AI security landscape has been jolted by a significant demonstration of vulnerability. The Snowflake AI system, a sophisticated multimodal model, was shown to have successfully…

The cybersecurity landscape is witnessing a paradigm shift with the introduction of BlacksmithAI, a novel open-source framework that positions artificial intelligence as the centra…