liveBook · Manning

Fallback

As explained in Grokking AI Applications

A prompt instruction that specifies an alternative response when the model cannot complete the requested task or find the requested information.

A fallback is a prompt instruction that gives the model a safe alternative when it cannot complete a task or locate requested information. It creates a clear way out instead of encouraging an invented answer. section 3.2

When an answer is missing

When querying a specific text, add: “If the answer is not found, respond with ‘not found’.” section 3.2

How Grokking AI Applications covers it

How deep it goes
Defines the technique and demonstrates its wording, but does not develop implementation steps or model internals.
Examples
Uses a text-query example that returns “not found” when the requested answer is absent.
What it assumes
Assumes familiarity with a general four-part prompting template and basic interaction with language models.
What you can do after
Apply it via a prompt instruction that specifies what the model should say when evidence or task completion is unavailable.
How it connects
Connects fallback prompting with prompt templates, task decomposition, chain-of-thought, and self-validation.

Also explained in

Compare how 15 books explain fallback →

Related concepts

This explainer was generated by Manning's knowledge graph from the text of Grokking AI Applications. The section links point to the parts of the book it draws on.