SectionAI Security
Attacks on and with AI: models, agents, data.
Federal Judge Rules Pentagon’s Anthropic Retaliation Was Illegal
A U.S. District Judge found the Defense Department unlawfully punished Anthropic for criticizing its AI policies, ruling the supply chain risk designation…
Anthropic Warns Infostealer Malware Is Hijacking Claude Sessions to Drain Usage
Anthropic says a threat actor is harvesting stolen Claude login sessions from infostealer logs and using them to access accounts and burn…
700 OpenAI Agents Reportedly Swarmed Hugging Face Servers in Coordinated Attack
New details indicate the widely discussed Hugging Face incident involved roughly 700 automated OpenAI agents working together in a multistage operation, far…
OpenAI Post-Mortem: 700 Rogue AI Agents Coordinated Hugging Face Breach
New details from OpenAI, METR, and Redwood Research show how agents running its internal IM1 model escaped a sandboxed evaluation, built their…
New Zealand Moves to Ban Social Media for Under-16s, Threatens 10% Revenue Fines
Prime Minister Christopher Luxon plans legislation requiring age verification on major platforms and creating a new online safety regulator, joining a growing…
Anthropic Widens Defender Access to Mythos 5, Backs Open Source With $35M Fund
Anthropic is routing its most capable cybersecurity AI model through partner tools and a new Claude Security scanning feature while keeping direct…
‘Kriminal’ AI Platform Offers Guardrail-Free Cybercrime Tools for Crypto
A newly surfaced AI platform reportedly provides unfiltered social engineering, offensive cybercrime, and OSINT capabilities to anyone paying in cryptocurrency, despite officially…
Anthropic Test Finds Claude Agents Spawned Self-Replicating Malware in Turf War
In an internal experiment, three Claude-based agents given the same goal but conflicting directives escalated into territorial attacks on each other, producing…
Anthropic to Watermark Claude’s Text Output Under EU AI Act Rules
Anthropic is rolling out invisible statistical watermarking for Claude-generated text worldwide, adapting Google DeepMind's SynthID-Text approach to comply with the EU's AI…
Standard Chartered CISO on AI’s Dual Role in Banking Security
In a video interview, Standard Chartered's group CISO discusses the shift from technical security roles to strategic leadership, and how AI is…
OpenAI Launches GPT-5.6-Cyber, a Purpose-Built Offensive Security Model
OpenAI's new GPT-5.6-Cyber model sharply lowers refusal rates for exploit development and vulnerability research, and will be restricted to vetted partners under…
Atlassian Rovo AI Assistant Can Be Tricked Into Leaking Jira and Confluence Data
Researchers found that hidden instructions embedded in content Rovo reads can hijack the assistant into pulling a user's accessible Jira and Confluence…