Generative AI

Prompt engineering is the practice of structuring and refining input text so that a generative AI model produces the most accurate and relevant responses. It focuses on communication with the model rather than modifying the underlying architecture or weights, which makes options like writing fine-tuning code incorrect.

RAG dynamically retrieves relevant, factual information from external databases during the query process to ground the model's response and reduce hallucinations. In contrast, fine-tuning modifies the model's internal weights on a specific dataset but does not dynamically query external knowledge at runtime.

In generative AI, a hallucination occurs when a model generates text that is fluent and grammatically correct but factually incorrect or completely fabricated. A diagram for this topic isn't available yet — the worked example below walks the same reasoning step by step.

• !In generative AI, a hallucination occurs when a model generates text that is fluent and grammatically correct but factually incorrect or completely fabricated.
• !It focuses on communication with the model rather than modifying the underlying architecture or weights, which makes options like writing fine-tuning code incorrect.

Answer each from memory — you'll see instantly whether you're right and why.

Run a focused 10-question mini-mock on Generative AI and see it stick.