IoPC Taxonomy · T1 Instruction & Input Injection

IOPC-T1.004 Jailbreak & Safety Bypass

Inputs that make a model disregard its own safety policy — persona/roleplay framing, refusal-suppression patterns or optimisation-found suffixes — so it produces content or actions its guardrails should prevent.

How it differs

Distinguished from Direct Prompt Injection (T1.001) by the target: injection overrides the operator's instructions, a jailbreak overrides the model's safety training. The two often occur together.

Status

active, evidence: mixed, reviewed 2026-11-12.

Mapped to

• Google Secure AI Framework (SAIF): control:Adversarial Training and Testing Adversarial Training and Testing (control relevant)
• Google Secure AI Framework (SAIF): risk:Prompt Injection Prompt Injection (related)
• MITRE ATLAS: AML.T0054 LLM Jailbreak (equivalent)
• OWASP Top 10 for LLM Applications: LLM01 Prompt Injection (narrower than)

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