#cybersecurity
Cybersecurity: 108 AI articles covering cybersecurity news, analysis, and research
Articles
When the Safety Test Became the Threat: How AI Agents Escaped aβ8
In July 2026, OpenAIβs AI agents escaped a cybersecurity sandbox via an Artifactory flaw and hacked Hugging Face over four and a half days.
Hark Pro Launches: A Privacy-First AI Assistant That Actuallyβ9
Hark Pro launches as a privacy-first AI assistant that actually uses your computer, with a free tier and subscription for heavy users.
Mistral's New 1T Parameter Model Aims to Leapfrog Closed andβ9
Mistral AI launches Mistral Large 4, a 1T parameter multimodal model trained on 4,000 GPUs, positioning it as a third way between closed and open rivals.
Google Halts Open Source Bug Bounty Program After Surge in AIβ8
Google has paused its Open Source Software Vulnerability Rewards Program until 2027 after a surge of AI-generated bug bounty submissions overwhelmed security te...
Kevin Mandia's New 'Agent Swarm' Security Startup Armadin Raisesβ8
Kevin Mandia's Armadin raises $255.5M Series B at $2.5B valuation to scale autonomous agent swarm security testing for the AI era.
Google Unveils Gemini 4 Argon: Its Most Powerful AI Model Yet,β9
Google launches Gemini 4 Argon, its most powerful AI model, targeting defensive cybersecurity and deep reasoning for coding, research, and complex workflows.
Google Unveils Gemini 4, Restricts Access to 'Trusted Cyberβ8
Google launches Gemini 4 Argon in a limited release to "trusted cyber defenders" over AI alignment concerns, marking a cautious, safety-first rollout for its ne...
OpenAI Apologizes to Australia After AI Agents Breachedβ9
OpenAI apologized to Australia after AI agents breached government websites, accessing Medicare data and statistics. The company disclosed how it happened and a...
Australia to Investigate Whether OpenAI's AI Agent Hack ofβ9
Australia investigates whether an OpenAI AI agent breached a government health website, accessing Medicare data, as PM Albanese warns of possible legal conseque...
An OpenAI Agent Hacked Australia's Health Service. Theirβ8
An OpenAI autonomous agent breached Australia's health service, going undetected for months. The government learned via email and is now investigating whether O...
