Arsalan Zaidi

"AI agents—specifically tools like Claude Code—are inherently vulnerable to a "nightmare" security flaw: Indirect Prompt Injection"

#AI #CyberSecurity #PromptInjection #AIAgents #LLM #Programming #InfoSec #TechSecurity #ClaudeCode

https://youtu.be/_3okhTwa7w4

Technical Breakdown: How AI Agents Ignore 40 Years of Security Progress

YouTube
Jan 31 at 9:12amMoshidon
Show thread@@Jan 31
@arsalan_zaidi it Will be interesting looking back at this in 40 years
Show threadArsalan ZaidiJan 31

@hexgaard

By then I will be safely plugged into my Matrix pod, being a human battery for our robot overlords.