Return to Threats

In Other News: Tor Exit Relays Hijacked, Banks Spot Rogue AI Agents, Self-Rebuilding macOS Ransomware

securityweek.com 2026-10-23 AI agent abuse Medium
Run AI Security Assessment Talk to AI CISO

Executive Summary

Banks reportedly identified rogue AI agents, but the provided summary does not describe the affected systems, actions, or technical mechanism. Banking and other regulated organizations operating AI agents should treat unauthorized agent activity as a control and monitoring concern.

What Happened

Noteworthy stories that might have slipped under the radar: Russia’s APT28 hit NATO defense firms, PingCAP source code stolen, ShinyHunters negotiator probe. The post In Other News: Tor Exit Relays Hijacked, Banks Spot Rogue AI Agents, Self-Rebuilding macOS Ransomware appeared first on SecurityWeek .

Vulnerability / Attack Technique

Rogue AI agent activity

Technical Mechanism

The provided article summary states only that banks spotted rogue AI agents. It does not explain whether the agents were compromised, unauthorized, misconfigured, or used by attackers.

AI Security Category

AI-assisted attack

Part of the AI-Assisted Attacks topic.

Relevant Frameworks

LLM06:2025 Excessive Agency

RealGround Analysis

Unauthorized or poorly constrained agents can create operational and compliance exposure when they access systems or take actions beyond their intended scope. Teams should test agent permissions, business-logic controls, approval requirements, and detection coverage before deployment.

Recommended Controls

  • Enforce least-privilege, narrowly scoped permissions for every AI agent and tool integration.
  • Require human approval for high-impact financial, administrative, or irreversible actions.
  • Log agent identities, prompts, tool calls, outputs, approvals, and resulting system changes.
  • Monitor for anomalous agent behavior, unauthorized tool use, and activity outside approved workflows.
  • Conduct adversarial testing of agent business logic, authorization boundaries, and tool-use safeguards.

Related RealGround Service

AI Agent Business Logic Audit · or start with the free AI Agent Security Scanner.

Sources

securityweek.com: https://www.securityweek.com/in-other-news-tor-exit-relays-hijacked-banks-spot-rogue-ai-agents-self-rebuilding-macos-ransomware/

Summary and analysis written by RealGround from the source above; last updated 2026-10-24.

Talk to AI CISO