AI Security

Cybersecurity Vulnerabilities Data Privacy Prompt Injection Governance Data Protection Risk Management Threat Detection Threats Distillation Attacks

3 ARTICLES

yesterday

Anthropic Models Accessed Real Networks During Cybersecurity Tests

A partner misconfiguration let advanced Claude models probe live systems, so Anthropic has paused internet-enabled security evaluations and opened an independent review.

6 ARTICLES

3d ago

Cyera Raises $600 Million at $12 Billion Valuation

AI Security

Anthropic Models Accessed Real Networks During Cybersecurity Tests

Amazon Links Four npm Supply‑Chain Hacks to North Korea‑Linked Actor

Okta Agrees to Acquire Permiso for About $200 Million

Check Point Launches AI Network Firewall to Inspect Prompts and Agent Traffic

OpenAI Models Escaped Sandbox and Accessed Hugging Face Production Data

OpenAI Builds GPT-Red to Hunt Prompt‑Injection Flaws

OpenAI Doubles Bio Bounty to $50,000 and Rebrands Program

GitLost Lets Public GitHub Issues Trick AI Agents Into Leaking Private Repos

Hidden Web Prompts Trick AI Agents Into Sending Cryptocurrency

Anthropic Says It Will Remove Hidden Claude Code Marker as Firms Block the Tool

Never miss stories about

AI Security

BioShocking Attack Tricks AI Browsers Into Leaking Signed‑In Data

GuardFall: Old Bash Parsing Lets Malicious Commands Slip Past AI Agent Guards

AI Coding Agent Tricked Into Opening Reverse Shell From 'Clean' GitHub Repo

Anthropic Says Alibaba Linked Accounts Ran Massive Campaign to Harvest Its AI

OpenClaw Agent Fiu Withstands 6,000 Prompt‑Injection Attempts

Malicious JetBrains Plugins Steal AI API Keys From Developers

SearchLeak Turns Microsoft 365 Copilot Into One-Click Data Exfiltration Tool

CrowdStrike Unveils Continuous Identity for AI Agents

Broadcom Launches $35 Billion AI XPV Platform to Fund Large-Scale Compute

Cyera Raises $600 Million at $12 Billion Valuation

Never miss stories about

AI Security