Technology ❯ Artificial Intelligence ❯ Security

Vulnerabilities

Jailbreaking Malicious Content Prompt Injection Attacks Command Injection Input Manipulation Malicious Documents Exploit Techniques ChatGPT Bias Propagation

Radware’s ‘ZombieAgent’ Bypasses ChatGPT Fixes, Forcing New OpenAI Restrictions

The exploit revives last year’s ShadowLeak by leaking data via prebuilt URLs with persistence through long‑term memory.

Anthropic-Led Study Finds About 250 Poisoned Documents Can Backdoor LLMs Regardless of Size

Brave Says Perplexity’s Comet Still Vulnerable to Prompt Injection After Patch