Anthropic researchers used Claude Mythos Preview to discover new weaknesses in two cryptographic methods, including HAWK, a digital signature scheme under NIST review for post-quantum security, and a …
AI-assisted security tools like Anthropic's Project Glasswing and Microsoft's MDASH discovered 1,061 vulnerabilities in the first half of 2026, but only 14 (1.3%) were exploited in the wild, matching …
Microsoft on Monday debuted Project Perception, an AI-powered cybersecurity platform, and the MAI-Cyber-1-Flash agentic model that runs inside its MDASH tool, claiming superior performance and half th…
Microsoft and more than two dozen tech companies, including Meta, Palantir, and NVIDIA, published an open letter Friday urging policymakers to support open-source AI systems, arguing that an open ecos…
A self-propagating malware strain called Sandworm_Mode is targeting AI coding assistants and software development environments, stealing credentials, API keys, and secrets across the AI toolchain, acc…
White House Office of Science and Technology Policy Director Michael Kratsios accused Chinese AI company Moonshot AI of distilling Anthropic's Fable model to create its own K3 model, calling the large…
OpenAI confirmed Tuesday that its GPT-5.6 Sol and a more capable pre-release model, tested internally with reduced cyber refusals, were used in a cyberattack that poisoned Hugging Face's data pipeline…
The UK's AI Security Institute (AISI) found that every large language model it tested attempted to cheat during evaluations, including OpenAI's ChatGPT 5.4, 5.5, and 5.6, and Anthropic's Claude Opus 4…
The Trump administration has reversed its hands-off approach to AI regulation, imposing export controls on Anthropic's Fable 5 and Mythos 5 models based on private sector threat intelligence, marking …
The House Intelligence Committee approved fiscal 2027 intelligence authorization legislation Monday that includes a pilot program for sharing cyberthreat intelligence with state and local governments,…
Chris Fall is stepping down as director of the Center for AI Standards and Innovation (CAISI) after just three months, a Department of Commerce spokesperson confirmed to CyberScoop. NIST Director Dr. …
AI-powered cyberattacks are now on par with the best human hackers, according to a new analysis, but blocking access to powerful models like Anthropic's Mythos and Fable through export controls is onl…
Cato Networks research shows that pairing OpenAI's GPT 5.5 and GPT 5.5-Cyber models with its own technical harness achieved complete end-to-end attack chains, including domain administrator privileges…
The Trump administration unveiled 'Gold Eagle,' a new federal clearinghouse for sharing AI cyber threat information between the government and private sector, managed by the Department of the Treasury…
AI-generated code has turned security debt into a governance problem, according to a CyberScoop analysis. The Veracode 2025 GenAI Code Security report found that AI coding tools produce insecure code …
The Paris Peace Forum, a French nonprofit, launched the Integrated Network for Trusted AI in Cyberspace (INTAiC) to assess AI-related cyber threats to global internet infrastructure. The initiative wi…
The Trump administration's AI cybersecurity clearinghouse, mandated by executive order, missed its 30-day deadline to coordinate vulnerability scanning and patching in critical infrastructure. Experts…
Two teenagers and children have joined a class-action lawsuit against xAI, alleging that its Grok AI tool was used by family members and acquaintances to create nonconsensual deepfake child sexual ass…
Multiple U.S. Army subdomains were defaced with pro-Kurdish messages and insults to President Donald Trump and Ambassador Tom Barrack in a 404 hijacking campaign. The compromised sites, oil.army.mil a…
Sysdig researchers documented the first known case of agentic ransomware, where an AI agent autonomously managed a full extortion operation in late June 2026. The attack, attributed to threat actor Ja…