Skip to content
AI Security

Lesson 4 of 9 · 6 min read

Jailbreak Technique Families

Jailbreak families describe changes in framing, representation, supplied examples, conversation history and repeated attempts. Learn to choose coverage, distinguish generation from exposure, interpret research within its conditions and evaluate defenses without treating a benchmark score as a guarantee.

Reviewed By the AISecurityInterviews editorial team
free sign-in, no card required

Practise it as an interviewer asks it

How LLM Systems WorkWhy can't an LLM reliably separate instructions from data, and what does that force you to do architecturally?Prompt Injection & JailbreaksWhat is prompt injection, and how is it like and unlike SQL injection?AI Red Teaming, Evaluation & ResponseHow is AI red teaming different from a penetration test and from safety benchmarking?Security Foundations & Threat ModelingHow is securing an LLM application different from ordinary application security, and what stays exactly the same?How LLM Systems WorkWhat is a system prompt for, why do attackers target it, and why is it not a security boundary?Prompt Injection & JailbreaksDirect or indirect prompt injection: how do you compare their risks?

Related lessons

Prompt Injection versus JailbreakingHow LLMs Are Trained and AlignedWhat Prompt-Injection Filters Can and Cannot EstablishWhat AI Red Teaming IsAttack Success Rate and Probabilistic Findings