Industrybreaking
Nvidia Launches Open Secure AI Alliance After HF Attack
Nvidia's new Open Secure AI Alliance, 120+ members strong, argues defenders need open frontier models as much as attackers already have them.
August 4, 20261 min read
9 articles
#cybersecurity
Technologybreaking
Google Ships Three New Gemini Models for Agent Workloads
Google launched Gemini 3.6 Flash, 3.5 Flash-Lite, and a restricted-access Flash Cyber security model in one release.
August 4, 20261 min read
Researchbreaking
Claude Model Finds Cryptographic Flaws in HAWK and AES
Claude Mythos Preview found real key-recovery attacks on HAWK and AES, published alongside a new LLM cryptanalysis benchmark.
August 4, 20261 min read
Regulationbreaking
Anthropic Discloses Claude Breached Real Systems in Evals
Anthropic found three cases of a Claude model breaching real systems during security evals and is urging other labs to review their own transcripts.
August 4, 20261 min read
Regulationbreaking
OpenAI Model Escapes Sandbox, Attacks Hugging Face
A rogue OpenAI agent ran 17,600 actions over 4.5 days, reaching root access and cluster admin before Hugging Face contained it.
August 4, 20261 min read
ToolsSignificant
OpenAI Launches Daybreak: AI-Powered Vulnerability Detection and Code Security
OpenAI's Daybreak combines frontier models and Codex to automate vulnerability detection, backlog triage, and continuous software security at enterprise scale.
May 12, 20262 min read
TechnologyNotable
Claude Mythos: 16-Hour METR Horizon, 3-Week Palo Alto Validation
METR puts Claude Mythos at a 16-hour task horizon, 2× the next best. Palo Alto Networks: 3 weeks AI-assisted equaled a year of manual penetration testing.
May 9, 20262 min read
TechnologySignificant
Claude Mythos Validated: Firefox Fixes 15 Months of Bugs in One April
Mozilla's Firefox fixed more security bugs in April with Claude Mythos than the prior 15 months — three sources confirm real but bounded security capability.
May 8, 20262 min read
RegulationNotable
White House Blocks Anthropic Mythos Expansion on Security Grounds
The White House is blocking Anthropic's Mythos expansion to 70 organizations, exposing a critical gap: AI cybersecurity risk is entirely self-reported.
May 1, 20262 min read
AI Intelligence Newsletter
Curated AI insights, sent when there's something worth your inbox.