Loading…
Frameworks and learning resources for building and defending AI systems securely.
Updated every few hours from The Hacker News, Dark Reading, Schneier on Security, Embrace The Red, Simon Willison and Google Security Blog
Teams are watching chainalysis confirms intrusion into internal tools and exposure of customer identity data as ai agents priorities shift in practice.
The Hacker News ·
Teams are watching a shared chatgpt link could run attacker code in your session and as security priorities shift in practice.
The Hacker News ·
Teams are watching chinese hackers use as security priorities shift in practice.
The Hacker News ·
Teams are watching official claude code extension on cursor marketplace trojanized with persistent ssh backdoor as llm priorities shift in practice.
The Hacker News ·
Teams are watching devs hit by trojanized claude code extension for cursor as llm priorities shift in practice.
Dark Reading ·
Teams are watching bloatware in aftermarket auto parts as llm priorities shift in practice.
Schneier on Security ·
Teams are watching threatsday as security priorities shift in practice.
The Hacker News ·
Teams are watching linux kernel tcp flaw with 9 8 cvss score affects ubuntu red as models priorities shift in practice.
The Hacker News ·
Teams are watching fake chatgpt and claude apps abuse signed executables to deploy crypto miners as llm priorities shift in practice.
The Hacker News ·
Teams are watching google s gemini 4 carbon leads benchmarks for finding exploiting and patching as llm priorities shift in practice.
The Hacker News ·
Teams are watching anthropic as security priorities shift in practice.
Dark Reading ·
Teams are watching rethinking cybersecurity awareness in the era as security priorities shift in practice.
Dark Reading ·
Teams are watching claude model builds exploits across major browsers oses as llm priorities shift in practice.
Dark Reading ·
Teams are watching shadowy relay disables openai codex claude code sandboxes as llm priorities shift in practice.
Dark Reading ·
Teams are watching takes root in hospital corridors raises security stakes as security priorities shift in practice.
Dark Reading ·
Teams are watching attacker relentlessly scans for credentials by impersonating bots as ai agents priorities shift in practice.
Dark Reading ·
Teams are watching 3 frameworks to protect cisos orgs against liability as security priorities shift in practice.
Dark Reading ·
Teams are watching hiding capabilities in models as models priorities shift in practice.
Schneier on Security ·
Teams are watching bias in qwen3 as models priorities shift in practice.
Schneier on Security ·
Teams are watching visual studio copilot and ssms remote code execution cve 2026 as security priorities shift in practice.
Embrace The Red ·
Teams are watching from select to sysadmin with sql copilot cve 2026 as security priorities shift in practice.
Embrace The Red ·
Teams are watching self as llm priorities shift in practice.
Simon Willison ·
Teams are watching breaking claude code opus 5 auto mode as llm priorities shift in practice.
Embrace The Red ·
Teams are watching recovering encrypted llm reasoning traces as llm priorities shift in practice.
Embrace The Red ·
Teams are watching stealing reasoning traces from proprietary llm apis as llm priorities shift in practice.
Simon Willison ·
Teams are watching auto mode is now the default in claude code for pro max as llm priorities shift in practice.
Simon Willison ·
Teams are watching incident report as llm priorities shift in practice.
Simon Willison ·
Teams are watching llm heist hijacking litellm for traffic interception key theft and tool as llm priorities shift in practice.
Embrace The Red ·
Teams are watching threats in the wild as llm priorities shift in practice.
Google Security Blog ·
Teams are watching google workspace s continuous approach to mitigating indirect prompt injections as llm priorities shift in practice.
Google Security Blog ·
Teams are watching vrp 2025 year in review as open source priorities shift in practice.
Google Security Blog ·
Teams are watching security for the quantum era implementing post as models priorities shift in practice.
Google Security Blog ·
Teams are watching staying one step ahead as llm priorities shift in practice.
Google Security Blog ·
Curated by the AI Hub Daily team
The most critical security risks for LLM apps, including prompt injection, sensitive data disclosure and excessive agency.
OWASP GenAI Security Project
A knowledge base of adversary tactics and techniques against AI systems, modelled on MITRE ATT&CK.
MITRE
Voluntary guidance for identifying and managing risks across the AI lifecycle: govern, map, measure, manage.
NIST
A game where you try to trick an LLM into revealing a password. A memorable way to see why guardrails are hard.
Lakera