Lesson 5 of 6 · 6 min read
Guardrails and Their Limits
Guardrails are controls placed around a model to check or constrain its inputs and outputs: schema validators, allowlists, PII detectors, topic and safety classifiers, injection detectors and LLM-based judges. Correctly implemented deterministic controls enforce their specified rules independently of model intent; probabilistic controls aim to lower the rate of bad outcomes and need honest measurement of both misses and false positives. A good architecture uses each where it fits and never presents a classifier as a boundary.
free sign-in, no card required
Practise it as an interviewer asks it
How is AI red teaming different from a penetration test and from safety benchmarking?How is securing an LLM application different from ordinary application security, and what stays exactly the same?Threat-model a customer-support chatbot that answers from a knowledge base and can look up orders.Explain tokens, context windows and sampling temperature, and why non-determinism changes how you write security tests.How do you evaluate a claim that prompt injection has been solved?Explain least privilege and defence in depth with an example where one layer failing does not cause a breach.