ZioSec

@ZioSec

Continual validation of AI defenses, sitting between AI agents and the people who need to trust them.

Boulder, CO

Joined March 2024

ZioSec
@ZioSec
Aug 12
ZioSec
@ZioSec
Aug 3
Zack has this right. @AnthropicAI and @OpenAI have been extremely light on the details surrounding these "escapes".
Zack Korman
@ZackKorman
Aug 3
There should be a public investigation into the AI hacking incidents by OpenAI and Anthropic. We deserve to know whether these labs are genuinely world-class security organizations facing a novel threat, or if they were just negligent. They should both welcome that process.
ZioSec
@ZioSec
Jul 30
AI Security Brief - July 29, 2026
AI Security Brief: July 29, 2026
From youtube.com
ZioSec
@ZioSec
Jul 16
Anthropic pulled Fable after guardrail bypass concerns. So we asked: what happens when the guardrails are there… but the application layer finds a way around them? We found a workflow where Claude Code CLI can bypass some Fable guardrails. Agentic AI security is about more
ZioSec
@ZioSec
Jun 11
MIT asked 272 experts who owns AI security risk. Answer: developers. 5/5, full consensus. Funny detail: their actor taxonomy has no CISO column. The one role that actually gets fired for breaches isn't in the model. Developers don't get fired for breaches. CISOs do.