Did it answer the words or help with the purpose?
Gricean relevance
Grice’s account of cooperative conversation includes the expectation that a contribution is relevant to the exchange. Context and implication matter alongside literal words.
An everyday example
You ask whether a train leaves at six because you need to reach a meeting. A useful response may also flag that the train arrives too late.
Original illustration, not an observed participant case.
Where the idea stops
This is not a rule that AI must answer only the literal question. A model’s claim that it “chose not to steer” is not evidence of its internal motives.
Something to try
State the purpose behind a question. Ask which relevant constraints remain unanswered, then verify the response against available facts.
Practise on a fictional case →Go deeper: the builder’s view
For builders: distinguish observed conversational behavior from a model-generated account of why it behaved that way.
Source and interpretation
This is a plain-language interpretation with an applied example; it is not a quotation or an endorsement from the source.
Grice — Logic and Conversation (1975) ↗Catalog v0.1 · checked 2026-10-29