Skip to content
OBLAIDISH NEWS
signal_tag · 2_broadcasts

#prompt-injection

// 2 transmissions tagged with #prompt-injection

LLMs hack custom vulnerable app in $1,500 test
TX_552886· AI

LLMs hack custom vulnerable app in $1,500 test

A developer built a deliberately insecure web app, spent $1,500 on API calls, and measured how well large language models could locate and exploit its flaws, revealing both promise and limits for AI‑driven security testing.

Microsoft Copilot's Cowork flaw lets attackers steal files via prompt injection
TX_753833· AI

Microsoft Copilot's Cowork flaw lets attackers steal files via prompt injection

A security flaw in Microsoft Copilot's Cowork feature allows file exfiltration through prompt injection, demonstrated by Kneenex on May 25, 2026 [hn-front].