{
  "id": 1732999,
  "title": "Portnox adds Microsoft Defender integration to police AI agent access",
  "url": "https://urgent.news/2026/08/18/portnox-adds-microsoft-defender-integration-to-police-ai-agent-access",
  "topic": "ai",
  "section": "AI",
  "published": "2026-08-18T13:00:05.000Z",
  "source": {
    "name": "SiliconANGLE",
    "slug": "siliconangle",
    "url": "https://siliconangle.com/2026/08/18/portnox-adds-microsoft-defender-integration-to-police-ai-agent-access/"
  },
  "original_language": "en",
  "account": "Cloud-native access control firm Portnox has introduced a new feature that integrates with Microsoft's Defender to manage AI agents and nonhuman identities accessing corporate networks. Portnox already supports CrowdStrike and SentinelOne integrations. These three platforms can now trigger access decisions based on threat detection or compliance issues. If a threat is reported, the Portnox policy engine can immediately restrict the connection, isolate the identity, or revoke its access altogether, depending on the customer's predefined rules. Employees are also subject to the same governance. These AI agents, unlike human employees, authenticate to systems, extract sensitive data, and perform actions autonomously and at a rapid pace. However, many companies still employ static credentials and limited permissions, often only verifying identity once upon login. This approach has led to security incidents, with 54% of enterprises reporting confirmed agent breaches and 69% admitting to shared credentials among agents. Portnox CEO Denny LeCompte emphasizes the need for continuous verification and governance of all identities connecting to enterprise networks. The company's solution acts as a \"kill switch\" at the network layer, allowing immediate restriction of access when trust is compromised, without waiting for an AI agent to escalate a security incident. The process involves three steps: detection of risk by one of the endpoint platforms, evaluation of the signal against customer access rules by Portnox's policy engine, and automatic enforcement. Field Chief Information Security Officer Garrett Gross highlights the disconnect between identifying an issue and taking action, noting that while many products can report anomalies, few can respond in real-time at the network level without manual approval. This integration provides security and IT teams with an audit trail, recording the identity, time, location, access permissions, and the policy that made the decision. Portnox, a venture-backed startup, has raised approximately $59.5 million, including a $37.5 million Series B round in April 2025.",
  "summary": "Cloud-native access control company Portnox today unveiled expanded capabilities aimed at governing the artificial intelligence agents and other nonhuman identities now logging in to corporate networks, applications and infrastructure. The new piece is an integration with Microsoft Corp.’s Defender. Portnox had already wired in CrowdStrike Holdings Inc. and SentinelOne Inc. Any of the three can…",
  "key_points": [],
  "editors_take": null,
  "illustration": null,
  "coverage": {
    "outlets": 1,
    "also_reported_by": []
  },
  "ai_generated": true,
  "disclaimer": "Summaries, key points and the editor’s take are written by software from other outlets’ reporting and may contain errors — always check the linked original."
}