liveBook · Manning

Ambiguity

As explained in Machines that Think

Ambiguity is the property of an expression or signal that permits multiple plausible meanings or interpretations.

Ambiguity arises when an expression permits more than one plausible interpretation. Words can have multiple meanings, sentences can be read in different ways, and intentions may be left unstated. Interpreting language therefore depends on context, world knowledge, and pragmatic inference—not just the words themselves. section 2.6.3

The meaning of “bank”

In translation, “bank” might refer to money or to a riverbank. Grammar and word form alone do not settle the choice; the surrounding context matters. section 2.2.2

Fixed patterns can produce plausible responses without resolving meaning. Template-based systems work within expected conversational forms, but need extensive handcrafted knowledge to choose reliably between alternatives. Early rule-based translation could also produce awkward or nonsensical results when it encountered ambiguous words. section 2.2.2 section 2.2.1

People draw on experience and context to resolve ambiguity, and dialogue can repair misunderstandings. Models instead infer from patterns in text: co-occurrence can reveal associations, but those associations are filtered through language rather than grounded in direct experience. A fluent output alone does not establish that the intended meaning has been recovered. section 2.6.3 section 2.2.2

How Machines that Think covers it

How deep it goes
Explains ambiguity across words, sentences, and unstated intentions, then contrasts human inference with models’ text-based associations.
Examples
Uses the financial and riverbank meanings of “bank” to illustrate context-dependent interpretation.
What you can do after
Evaluate how fixed patterns and text-based associations can leave intended meaning unresolved.
How it connects
Connects ambiguity to context, pragmatic inference, pattern matching, and machine translation.

Also explained in

Compare how 6 books explain ambiguity →

Related concepts

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