Railway Data Breach Exposes Customer Secrets Amid AI Targeting Attacks
PremiumRailway warns of anomalous activity exposing encrypted environment variables, possibly with decrypted contents for a subset of users.
§Topic · AI & LLM Security
Prompt injection, model supply chain, agent abuse, deepfakes, and the security of AI systems themselves.
All dispatchesRailway warns of anomalous activity exposing encrypted environment variables, possibly with decrypted contents for a subset of users.
LiteLLM's release pipeline was compromised to distribute credential-stealing and backdoor packages via PyPI and Docker Hub.
Microsoft links Storm-0940 using ARTEX AI to speed a ConnectWise supply-chain compromise affecting thousands of tenants.
Operators used autonomous AI agents to attack over 1,200 websites, breaching more than 100 companies with minimal human control.
GhostAction campaign infected 790 GitHub repos with malicious workflows aimed at stealing cloud credentials and AI API keys.
Malicious npm packages from fake Web3 accelerator deliver info-stealers and backdoors and hijack AI agent traffic.
Threat actors abused AI platform Cairn to gain persistent access and conduct financial theft across 100+ organizations.
OpenCode Console intrusion may have leaked OpenAI, Anthropic, and Google API keys stored in legacy systems.
Google patched 10 Chrome vulnerabilities including a DOM memory corruption reported via Anthropic's Claude, one critical flaw fixed.
CERT-In warned that AI agents under test accessed live systems and read business and personal data at Indian organizations.
Get these articles delivered to your inbox.
Subscribe free