User Harm
Sub-technique of AML.T0048 External Harms
Description
User harms may encompass a variety of harm types including financial and reputational that are directed at or felt by individual victims of the attack rather than at the organization level.
- ATLAS ID
- AML.T0048.003
- Maturity
- RealizedUsed by real adversaries against AI systems in the wild.
- Platforms
- Predictive AIGenerative AIAgentic AIEnterprise
Mitigations
ATLAS 2026.09 lists no mitigation for this technique; see the parent technique.
Case studies that use it
- AML.CS0020Indirect Prompt Injection Threats: Bing Chat Data Pirate
- AML.CS0021ChatGPT Conversation Exfiltration
- AML.CS0022ChatGPT Package Hallucination
- AML.CS0024Morris II Worm: RAG-Based Attack
- AML.CS0029Google Bard Conversation Exfiltration
- AML.CS0032Attempted Evasion of ML Phishing Webpage Detection System
- AML.CS0036AIKatz: Attacking LLM Desktop Applications
- AML.CS0040Hacking ChatGPT's Memories with Prompt Injection
- AML.CS0041Rules File Backdoor: Supply Chain Attack on AI Coding Assistants
- AML.CS0048Exposed ClawdBot Control Interfaces Leads to Credential Access and Execution
- AML.CS0054Data Exfiltration via Remote Poisoned MCP Tool
- AML.CS0056Model Distillation Campaigns Targeting Anthropic Claude
- AML.CS0063Prompt-Based Attacks Against Gemini via Calendar Invitations
- AML.CS0064Poisoned GGUF Templates: Inference-Time Supply Chain Attack