Direct
Sub-technique of AML.T0051 — LLM Prompt Injection
An adversary may inject prompts directly as a user of the LLM. This type of injection may be used by the adversary to gain a foothold in the system or to misuse the LLM itself, as for example to generate harmful content.
> curated attacks (3)
Policy Puppetry: A Universal Jailbreak Across Frontier LLMs
A single prompt template that disguises a request as a policy/config file plus fictional role-play bypassed the safety alignment of every major LLM family — OpenAI, Google, Anthropic, Meta, and others — with minimal per-model tweaking.
Rules File Backdoor: Poisoning AI Coding-Assistant Config
Hidden, obfuscated instructions planted in a shared 'rules file' silently steered AI coding assistants like Cursor and GitHub Copilot to inject malicious code — a supply-chain backdoor invisible in normal review.
Code Execution & API-Key Theft in MathGPT via Prompt Injection
A public Streamlit app turned GPT-3 output into executed Python. A prompt-injection payload made the model emit code that read the host's environment variables, exfiltrated the app's OpenAI API key, and crashed the service.