OpenAI patents setting that lets users choose how much effort AI spends on answers

1. A computer-implemented method comprising: presenting, by a front end to a generative response engine, a prompt composer interface with an indicator for a compute allocation user interface that previews two or more selectable compute allocation settings that affect a generation of a response by the generative response engine; upon selecting of the indicator, presenting the compute allocation user interface in the prompt composer interface wherein the compute allocation user interface includes a selector object that is configured to be manipulated by a pointer device to indicate a compute allocation setting selected from the two or more selectable compute allocation settings that include a default compute allocation setting, at least one greater compute allocation setting, or at least one lesser compute allocation setting; receiving, from a user account and by the front end to the generative response engine, a prompt and a selection of the compute allocation setting in the compute allocation user interface, wherein the compute allocation setting controls a relative amount of tokens that the generative response engine can generate when generating the response to the prompt; receiving, by the generative response engine, the prompt and the compute allocation setting associated with the selection of the compute allocation setting; and generating, by the generative response engine, the response to the prompt based on the compute allocation setting indicated by the selection of the compute allocation setting in the compute allocation user interface, wherein the at least one greater compute allocation setting causes the generative response engine to use more tokens than the default compute allocation setting when generating the response to the prompt and the at least one lesser compute allocation setting causes the generative response engine to use less tokens than the default compute allocation setting when generating the response to the prompt.