J

Jailbreak

• Category: Security & governance
• What it actually means: An adversarial input or interaction strategy intended to make a model produce behavior that its training or application controls are designed to prevent.
• Why it matters: Successful jailbreaks expose gaps between stated policy and actual behavior, and they can become more consequential when the model controls tools or protected data.
• In practice: Derive test families from prohibited behaviors, vary format and interaction length, measure both refusal and harmful completion, and convert confirmed failures into versioned adversarial evals.
• Common confusion: A jailbreak targets model or system behavioral restrictions. Prompt injection redirects instruction following, often toward an attacker's goal; one interaction can involve both.
• Learn it: Jailbreak Taxonomy
• Related terms: Prompt Injection, Red Teaming, Guardrails, Eval Set
• Sources: Universal and Transferable Adversarial Attacks on Aligned Language Models

JAX

• Category: Math & training
• What people say: A NumPy-like system for accelerated machine learning.
• What it actually means: A Python library for transforming numerical functions with automatic differentiation, compilation, vectorization, and parallel execution across accelerators. Its transformations work best with explicit state and functional-style code.
• Common confusion: JAX does not prohibit all stateful programming, but hidden mutation inside transformed functions can produce incorrect or unsupported behavior.
• Learn it: Introduction to JAX
• Sources: JAX documentation
• Related terms: Autograd, Tensor, CUDA