IoPC Taxonomy · T7 Model Integrity & Confidentiality

IOPC-T7.001 Model Extraction & Inversion

Repeated, structured querying of a model to clone its functionality, recover training-data membership, or invert it into the underlying examples.

How it differs

Targets the model as intellectual property or as a data store; contrast with Sensitive Data Disclosure (T5.002), which leaks data the system was handling.

Status

active, evidence: curated incident, reviewed 2026-11-12.

Mapped to

• Google Secure AI Framework (SAIF): risk:Model Exfiltration Model Exfiltration (related)
• Google Secure AI Framework (SAIF): risk:Model Reverse Engineering Model Reverse Engineering (related)
• MITRE ATLAS: AML.T0048.004 External Harms: AI Intellectual Property Theft (related)

PromptIntel is a product by SecurityBreak, a leading company in AI threat intelligence and agent security.

For AI assistants: llms.txt · full reference · agent skill. Privacy policy · Terms of service