Jerusalem, Israel
2017
  |  By Ben Hirschberg
Every guide to AI agent observability tells you what to capture — prompts, tool calls, token usage, traces, syscalls. Almost none address which of those signal sources you can still trust when the agent itself is part of the threat model. That distinction is the entire difference between observability that helps your SRE team debug a slow reasoning chain and observability that helps your security team investigate a breach.
  |  By Shauli Rozen
When AI agent workloads start generating more alerts than your SOC can keep up with, the instinct most teams reach for is to deploy more triage on top of what they already have. If the SIEM is producing thousands of atomized alerts, plug in something downstream that can cluster, prioritize, and auto-resolve them faster than a human can. The market has consolidated around exactly this answer.
  |  By Yossi Ben Naim
At 2:47 PM on a Tuesday, a customer support agent receives a routine ticket asking about return policy edge cases. The agent retrieves a section from your internal policy wiki through RAG to formulate the response. Three weeks earlier, an attacker had planted a hidden instruction in that wiki page. Bedrock Guardrails scored the retrieved context at 0.04 — well within benign range.
  |  By Ben Hirschberg
MITRE ATLAS catalogs sixteen tactics and eighty-four techniques adversaries use against AI systems, including fourteen agent-focused techniques added through the October 2025 Zenity Labs collaboration. It is the canonical taxonomy a security architect’s CISO, auditor, or RFP will name. It is not a detection plan. ATLAS organizes around adversary objectives.
  |  By Shauli Rozen
It usually starts the same way. The CISO comes back from a board meeting having signed off on agentic AI for production. The SOC lead is told, in roughly that many words, to build detection for the agents. And the security stack she has — CNAPP for posture, EDR on the nodes, container runtime sensors, a SIEM ingesting everything — was architected before AI agents existed as a workload class.
  |  By Yossi Ben Naim
A platform team finishes a two-week observation window on a new internal research agent. The baseline is stable; the sensor produced a clean profile. By Friday, no policy has shipped — and the blocker isn’t tooling.
  |  By Yossi Ben Naim
The residency evidence GDPR and the EU AI Act now expect lives in the runtime trajectory of every AI agent execution, not in the deployment configuration. Your residency compliance dashboard — every workload in eu-west-3, sovereign cloud configured, SCCs signed — cannot produce it. Your AI agent’s last thousand inferences crossed an external border, on average, eight times each. The translation API routed through us-east-1 when the EU endpoint hit capacity.
  |  By Ben Hirschberg
Editing IAM policies cannot fix the most common architectural mistake in shipping AI agents on Kubernetes. It happens in thirty seconds: a platform engineer reuses an existing ServiceAccount with an IRSA annotation for Bedrock access because creating a new one takes thirty minutes plus a Terraform pull request. The new agent ships under the existing identity.
  |  By Shauli Rozen
Your risk committee meets Thursday. The agenda has a new item: AI agent risk posture. You open the register. The fraud detection agent shipped in March is on it. So is the customer service agent. Neither row is useful — “likelihood: medium, impact: high, control: service account scoped via IAM.” Three months ago that was approximately right. Last week the platform team added two MCP connections, the model was upgraded, and the agent now touches data classes the entry never anticipated.
  |  By Shauli Rozen
Most “hardening” advice for AI agents is a checklist of things to configure before the agent runs. CIS Kubernetes Benchmark gates. Pod Security Standards baselines. NetworkPolicy templates. None of it’s wrong — it’s just one of four phases, the one your stack already covers. The other three are Observe, Enforce, and Reconcile. They’re where AI agents actually get breached, and they’re where most stacks have nothing.
  |  By ITProTV
With the short week for the Thanksgiving holiday in the US, the Technado team decided to have a little fun by looking back at some of the dumbest tech headlines from 2019. Romanian witches online, flat-earthers, and fake food for virtual dogs - what a time to be alive. Then, Shauli Rozen joined all the way from Israel to talk about a zero-trust environment in DevOps. IT skills & certification training that’s effective & engaging. Binge-worthy learning for IT teams & individuals with 4000+ hours of on-demand video courses led by top-rated trainers. New content added daily.

ARMO closes the gap between development and security, giving development, DevOps, and DevSecOps the flexibility and ease to ensure high grade security and data protection no matter the environment – cloud native, hybrid, or legacy.

ARMO is driving a paradigm shift in the way companies protect their cloud native and hybrid environments. We help companies move from a “close-the-hole-in-the-bucket” model, installing firewalls, defining access control lists, etc. to a streamlined DevOps- and DevSecOps led model in which environments are deployed with inherent zero-trust.

Security at the Speed of DevOps:

  • Runtime workload identity and protection: Identifies workloads based on application code analysis, creating cryptographic signatures based on Code DNA to prevent unauthorized code from running in the environment to access and exfiltrate protected data. The patent-pending technology signs and validates workloads in runtime throughout the entire workload lifecycle.
  • Transparent data encryption: Transparent data encryption – keyless encryption – robustly and uniformly encrypts and protects files, objects, and properties, requiring no application changes, service downtime, or impact on functionality. It eases the adoption of encryption by removing the complexity of key management and providing an out-of-the-box solution for key protection in use, key rotations, and disaster recover procedures.
  • Identity-based communication tunneling: Transparent communication tunneling ensures only authorized and validated applications and services can communicate. Even if attackers steal valid access credentials, they are useless because the malicious code will be unsigned. Create API access polices to build identity-based policies and enforce correct workload behaviors.
  • Application-specific secret protection: Application-specific protection of secrets ensures cryptographic binding between continuously validated specific workload identities and their confidential data, delivering complete protection against access by unauthorized applications.
  • Visibility & compliance: Visibility and compliance monitoring provide granular details about workloads and running environments, including individual processes, file names and locations, open listening ports, actual connections, mapped volumes, opened files, process privilege levels, connections to external services, and more. Alerts can be used for continuous compliance verification.

Bringing Together Run-Time Workload And Data Protection To Seamlessly Establish Identity Based, Zero-Trust Service-To-Service Control Planes.