AI Tools Best Practices Vocabulary

What we verified before publishing

Fact-check status: passed

The following claims were verified through Perplexity Sonar before the questions were finalised. The host can reference any of them during play:

• A prompt is an input request, such as a command or question, used to guide an AI system to generate a specific response or perform a task.
• A large language model (LLM) is a type of neural network trained on vast amounts of text data to understand, summarize, translate, and generate human-like language.
• AI hallucination occurs when a model generates a response that is factually incorrect, nonsensical, or fabricated while being presented as a confident, plausible fact.
• The context window is the maximum amount of information, measured in tokens, that an AI model can process and reference during a single interaction.
• Zero-shot prompting is a technique where an AI is asked to perform a task without being provided any specific examples of the desired output.
• Few-shot prompting involves providing an AI model with a small number of input-output examples within the prompt to guide its performance on a specific task.
• Retrieval-augmented generation (RAG) is an AI framework that improves response accuracy by retrieving relevant information from external, authoritative knowledge bases before generating an answer.
• Temperature is a hyperparameter that controls the randomness of an AI model's output, where lower values produce more deterministic, focused responses and higher values increase creativity.
• Fine-tuning is the process of taking a pre-trained foundation model and further training it on a smaller, task-specific dataset to improve its performance in a particular domain.
• Prompt engineering is the practice of designing, refining, and optimizing prompts to guide AI models toward generating more accurate, relevant, and useful outputs.