ThreatFabric Announces Agentic AI Risk and Know Your Agent Capabilities

Summary
ThreatFabric says it now supports Agentic AI Risk and Know Your Agent capabilities, combining device, behavioral and session intelligence to assess agentic fraud risk. It reports finding more than 10 AI agent apps on customer devices running banking apps.
Key points
- ThreatFabric says it found more than 10 AI agent apps on devices running its customers’ banking apps.
- According to ThreatFabric, most of those apps were installed outside Google Play and used Android Accessibility Services to read screens and interact with apps.
- ThreatFabric’s announced approach combines device intelligence, behavioral analytics and session monitoring to assess agentic risk before transactions.
- The company says its capabilities aim to distinguish humans, agents and bots, and to assess agent behavior, context and intent.
- The article identifies risks including hijacked agents, agent-driven scams, agentic account takeover and legitimate agents performing malicious actions.
- Google announced a beta of its on-device AppFunctions framework for Android and Computer Control for UI-based automation.
Article Details
- Event Type
- ThreatFabric announced support for Agentic AI Risk and Know Your Agent (KYA), capabilities for assessing AI-agent identity, behavior, context, and intent across mobile and web banking channels.
- Impact
- ThreatFabric reports finding more than ten AI-agent apps on devices running its customers’ banking apps; most were installed outside Google Play and used Android Accessibility Services to read screens and interact with apps. The article warns that agents may be hijacked or used for malicious actions, creating risks including device takeover, scams, account takeover, and fraudulent account opening. It also cites Interpol’s claim that AI-enhanced fraud is 4.5 times more profitable for criminals.
Vendors
Products
Agentic AI RiskThat’s why ThreatFabric now supports Agentic AI Risk and Know Your Agent (KYA): distinguishing humans from AI agents, legitimate agents from malicious ones, and a new AI model to differentiate good intent from bad,AndroidGoogle has announced the beta of its on-device AppFunctions framework for Android, alongside Computer Control for UI-based automation.AppFunctionsGoogle has announced the beta of its on-device AppFunctions framework for Android, alongside Computer Control for UI-based automation.Computer ControlGoogle has announced the beta of its on-device AppFunctions framework for Android, alongside Computer Control for UI-based automation.Know Your Agent (KYA)That’s why ThreatFabric now supports Agentic AI Risk and Know Your Agent (KYA): distinguishing humans from AI agents, legitimate agents from malicious ones, and a new AI model to differentiate good intent from bad,
