In this episode, host Andrius Useckas is joined by Aaron Walls and Alex Gatz to break down the explosive growth of Open Claw in enterprise environments — and the security nightmares that come with it. 

Plus, a special conversation with Isaac Qureshi, Co-Founder & CEO of Gatlin Robotics, on what happens when AI agents meet the physical world.

Topics covered:
🔒 Enterprise Open Claw Adoption — With 22% of enterprises already running Open Claw (often without IT's knowledge) and 40,000+ exposed instances, the team digs into why banning it doesn't work and what CISOs should actually do about it.
🛡️ Iron Claw & Secure Alternatives — Aaron shares his hands-on experience with Iron Claw's web assembly sandboxing approach. The verdict? More secure by design, but so restrictive it loses what makes Open Claw useful in the first place.
💉 Prompt Injection Epidemic — HackerOne reports a 540% increase in prompt injection attacks in 2025, with only 26% getting mitigated. The group debates whether model providers even have incentive to fix this — and whether regulation will force their hand.
⚖️ Regulation vs. Innovation — From the EU AI Act to Colorado's failed legislation and NIST's open calls for comment, the team discusses why compliance frameworks (PCI, HIPAA) haven't caught up and whether early regulation kills innovation.
🤖 Robotics + AI Agents (feat. Isaac Qureshi) — Isaac walks through Gatlin Robotics' approach to building cleaning robots with human-in-the-loop AI, the real risks of prompt injection via physical inputs (like writing on a whiteboard), and why maintaining a "knowledge gap" between human and AI is critical.
🧑‍💻 AI Agents Hiring Humans — The dystopian-sounding but very real marketplace where Open Claw agents can task humans to complete physical-world actions. TaskRabbit, but your boss is an AI.
🔮 Where Robotics + Agents Are Headed — From Pico Claw on Raspberry Pi to humanoid fleet systems, the conversation closes on how fast this space is moving and why security can't afford to be an afterthought.

🎙️ Along The Edge — AI security topics that matter, from the people working on the front lines.

Along the Edge — Episode 3

How do you break an AI agent? Javi Rivera — AI security researcher at ZioSec with 8+ years of offensive security experience from MITRE to ThreatX — breaks down the real-world techniques attackers use against agentic AI systems.

In this episode, we cover:

• Jailbreaks vs. prompt injections — what's the actual difference and why it matters
• Why classic attacks still work — SQL injection, command injection, and XSS through AI agents as a "middleman"
• System prompt extraction — how attackers use leaked instructions to craft targeted exploits
• MCP server security — why public MCP catalogs are the new supply chain risk and why there's no good solution yet
• Validating real findings vs. hallucinations — the hardest problem in AI pentesting
• Live demo — Gray Swan arena walkthrough showing indirect prompt injection in action
• Defense strategies — least privilege, sandboxing, guardrails, and why defense in depth still applies
• The coming threat — nation-state AI agents, automated offensive tooling, and why the next wave of attacks will be unprecedented

Whether you're a red teamer, AI developer, or security leader deploying agentic AI — this is the technical deep dive you need.

Resources mentioned: Gray Swan AI Arena, HackerPrompt, NVIDIA NeMo Guardrails, Docker MCP Hub

OpenClaw (formerly Clawdbot / Moltbot / whatever it’s called today) is the first agent that feels like “Siri, but real” — and it’s moving so fast it’s breaking everyone’s threat models in real time.

In this episode of Along The Edge, we unpack why OpenClaw is blowing up, what it can do when you hook it into your email, calendar, code, and tools… and why the security tradeoff is brutal: the more capable it is, the more dangerous it becomes.

We cover:

• Why “credentials in cleartext” is just the beginning
• How Discord / chat integrations can leak gateway + session details
• Tool invocation endpoints and bypass paths
• MCP prompt injection turning “normal workflow” into command execution
• What attackers will fingerprint and scan for in the wild
• What CISOs should do on day 1
• The big question: can defense keep up, or do we go “offense-driven defense”?

Buckle up.

Welcome to Along The Edge, a podcast about AI security and agentic AI.

In Episode 1, Andrius Useckas (Co-founder & CTO, ZioSec) sits down with Alex Gatz (Staff Security Architect, ZioSec) to break down the emerging world of agentic AI security: jailbreaks, prompt injection, SDR and SOC agents, data leaks, least privilege, and why “don’t worry, the model will filter it” is a dangerous assumption.

They also walk through V-HACK, an intentionally vulnerable agentic lab project that lets security researchers and pentesters safely experiment with agent exploits, tool calling, jailbreaks, and attack paths—helping define what “pen tester 2.0” looks like.

Chapters / In this episode:

00:00 – Intro: who we are & why a new AI security podcast
02:00 – What is agentic AI vs a plain LLM?
03:10 – SDR agents, SOC workflows & new “Layer 8 / Layer 9” problems
09:00 – Prompt injection 101: direct vs indirect attacks & context windows
12:00 – Chatbots vs agents and why agent risk is higher
15:00 – Foundation model trust & the Anthropic horror-story jailbreak demo
19:30 – Why jailbreaks are (currently) an unsolved problem
22:30 – Social engineering parallels & detecting AI / agentic attacks
27:00 – V-HACK: intentionally vulnerable agent lab for pentesters
32:00 – Securing agents: WAFs, runtime protection, identity & MCP proxies
36:00 – Scanners, evals vs real pentesting & terrifying token bills
39:00 – Least privilege, DLP & identity for SDR and payroll-style agents
44:00 – “Don’t trust, verify”: threat modeling & testing agents early
46:00 – Future of AI security: consolidation, CNAPs & SOC-as-an-agent
49:00 – Magic wand: fixing context & memory in agents
50:30 – Closing thoughts & what’s next

Links mentioned:

ZioSec – www.ziosec.com
V-HACK (GitHub) – https://github.com/ZioSec/VHACK

About the guests:

Andrius Useckas has 25+ years in security and now focuses on agentic AI security, offensive testing, and red teaming for enterprise AI deployments.

Alex Gatz is a Staff Security Architect at ZioSec. He has a background in emergency medicine and construction, then transitioned into AI in 2014 working on NLP, deep learning, anomaly detection, and now AI security.

If you’re building or testing agents in 2026, this episode gives you a practical look at how real attack paths work, what breaks in production, and how to defend before attackers get there first.