Cyber Attacks
Phishing emails now target AI assistants with hidden prompt injection instructions
Attackers embed concealed commands in emails to manipulate AI inbox summarizers into generating fraudulent to-do items for recipients.
prompt injection
Everything Threat Times has reported about prompt injection: 4 stories, newest first. Part of our Vulnerabilities coverage.
AI
Prompt Injection Warning Signs: How to Spot AI Manipulation
Many prompt injection attacks hide inside normal-looking text, bypassing surface-level security checks through subtle instruction overrides.
Vulnerabilities
Praisonai AICoder vulnerability allows arbitrary file write and command execution
A critical flaw in Praisonai’s AICoder component enables attackers to write files anywhere on the filesystem and execute commands via prompt injection.
Vulnerabilities
PraisonAI Agents CVE-2026-61447 Allows Full Environment Secret Exfiltration
A critical vulnerability in PraisonAI Agents allows attackers to steal API keys and execute arbitrary code by bypassing sandbox controls through LLM-generated Python scripts.