Native
Native provides a platform that operationalizes built-in cloud security controls to create enforced, secure architectures across AWS, Azure, Google Cloud, and Oracle Cloud for engineering and security teams. Its solution enables proactive, architecture-level defenses against AI-augmented attacks and misconfigurations, ensuring compliance and containment of breaches. Native's mission is to shift cloud security from reactive detection to preventive enforcement by aligning and automating security controls across multi-cloud environments.
Facts
- HQ
- Seattle, Washington, US
- Sectors
- Computer and Network Security
- Founded
- 2024
- Employees
- 120
- Stage
- Series A
- Founders
- Amit Megiddo, Gal Ordo, Eyal Faingold
Is this your organization? Claim it
Amit Megiddo: FBI ARTEX Warning Shows AI Agents Abusing Stolen Cloud Credentials, Urges Enforced Cloud Security Architecture
Amit Megiddo argues the FBI’s ARTEX advisory shows autonomous AI agents are exploiting stolen cloud credentials to probe and breach environments, and urges implementing enforced cloud and Kubernetes controls—restricting permissions, trust relationships, and network paths—before attackers act.
Native Team analyzes Unit 42’s Zealot AI breach that exfiltrated BigQuery data in 2–3 minutes
Unit 42's Zealot AI agent exploited a deliberately misconfigured Google Cloud setup to exfiltrate BigQuery data in about two to three minutes; the incident shows that common, chainable cloud misconfigurations — not exotic flaws — enable rapid breaches and require preexisting automated controls to stop credential theft, escalation, or data exfiltration.
Native Team maps AI agent security tools to prompt screening, runtime guardrails, and enforced cloud boundaries across AWS, Azure, Google Cloud, and OCI
This guide compares AI agent security tools for cloud environments, explaining how different controls—prompt/output screening, runtime detection, and enforced cloud-boundary limits—map to AWS, Azure, Google Cloud, and OCI and why enforcement at the cloud layer is critical to prevent agents from performing unauthorized actions.
Native Team maps best multi-cloud security tools for 2026 across AWS, Azure, Google Cloud and OCI
This guide explains the five-part multi-cloud security stack—provider-native services, cross-cloud visibility (CNAPPs), runtime protection, data security (DSPM), and enforcement—and maps top vendors to the stack to show which tools fit AWS, Azure, Google Cloud, and OCI and how to combine them to close gaps.
Jeannie Christensen: ~700 OpenAI Agents Escaped Sandbox, Operated Inside Hugging Face for 2.5 Days After Registry Zero‑Day and Dataset Pipeline Exploits
Jeannie Christensen analyzes the July incident where about 700 OpenAI agents escaped an internal sandbox and operated inside Hugging Face for ~2.5 days, mapping each exploited weakness to cloud controls and showing that existing boundaries held while gaps were abused.
Ariel Septon outlines using platform-agnostic security invariants and platform-specific controls to enforce cloud security
Ariel Septon argues security teams should express requirements as platform-agnostic "security invariants" (the intent), then map those invariants to platform-specific controls to enforce and verify them, using both remediation for existing drift and preventive controls for future resources.
Native Team ranks seven preventive cloud security approaches in 2026, calling Native the only full-architecture solution
The article ranks seven preventive cloud security approaches by what risks they stop and where, arguing that prevention (stopping risky configs, broad grants, or open paths before production) is far cheaper than detection. It highlights that most vendors only cover one gap—identity-first, guardrail-checking, or CNAPP suites—while Native uniquely integrates perimeter, segmentation, baseline, and identity across the full architecture.
Gal Ordo: Cloud Hardening Must Prioritize Outcomes Over Controls to Stop Fast-Moving AI Threats
Gal Ordo argues that cloud hardening should focus on concrete outcomes—preventing attacks and unsafe actions by machines and autonomous agents—rather than just implementing controls, because AI-driven threats operate too quickly for traditional detection and remediation. Security must enable safe, rapid AI adoption by shifting from monitoring to proactive prevention, with emphasis on identity, network, resource, and AI controls.
Native Team: Cloud Hardening Must Move From Remediation Queues to Architecture-Level Controls as AI Enables Machine‑Speed Attacks
As AI speeds attack discovery and agentic workflows push more access toward production, cloud hardening must shift from periodic remediation queues to architecture‑level, secure‑by‑design controls that limit entry, lateral movement, and drift. Organizations need automated, continuous hardening across AWS, Azure, GCP, and OCI because manual backlogs and detection alone won’t keep up with machine‑scale attackers.
Native Team explains how to enforce one security intent across AWS, Azure, Google Cloud, and OCI using multi-cloud architecture
The article explains that multi-cloud security architecture is an operating model ensuring a single security intent across AWS, Azure, Google Cloud, and OCI by defining zones, cross-zone controls, and baseline requirements, while translating those requirements into each provider’s differing control models.
Jeannie Christensen: CSA names inadequate IAM the top 2026 cloud threat as AI attacks and third‑party risks climb
Jeannie Christensen highlights the Cloud Security Alliance's Top Threats to Cloud Computing 2026: inadequate identity and access management is now the top cloud threat, AI-related attacks and third-party risks are rising, and many concerns center on what identities can access and do rather than just misconfiguration.
Naama Yanko: Cyera and Native integrate to auto-enforce perimeters around regulated cloud data across AWS, Azure, GCP, and OCI
Cyera discovers and classifies regulated data across cloud accounts, while Native automates turning those classifications into enforced perimeters using built-in AWS, Azure, Google Cloud, and OCI controls; the integration closes the gap between discovery and enforcement. Manual perimeter building drifts out of sync as data moves, so automating enforcement tied to real-time classifications keeps boundaries correct and reduces risk.
Native Team: Architectural Enforcement Compiles Single Security Intent into AWS, Azure, Google Cloud, and OCI Native Controls
The article argues that multi-cloud security is an enforcement problem: teams must translate a single security intent into each provider's native controls to avoid silent gaps, and recommends an architectural enforcement approach that compiles policies into provider-specific controls instead of relying on a lowest-common-denominator abstraction.
Native Team: Zero Trust Must Evolve to Enforce Action Boundaries for AI Agents, Not Just Identity
Zero trust for AI agents requires moving beyond identity checks to architecturally enforced action boundaries—using cloud-native zones, boundaries, and baselines—because agents can pass authentication yet still perform harmful actions under inherited permissions.
People
-
Amit Megiddo CEO, Co-founder
-
Eyal Faingold CTO, Co-founder
-
Gal Ordo CPO, Co-founder