Detecting compatibility mismatch by generative artificial intelligence

A system comprising:

• one or more processors;
• and
• a computer-readable storage medium storing instructions that are executable by the one or more processors to perform operations comprising:
• retrieving compatibility data from a database, the compatibility data comprising a list of recommended compatibilities between a plurality of items and a plurality of categories of vehicles and a first user reported compatibility between at least one item of the plurality of items and at least one category of vehicle of the plurality of categories of vehicles;
• training, by an artificial intelligence model manager and using at least a portion of the compatibility data, a machine learning model to detect a compatibility mismatch between a category of vehicle and an item, wherein the machine learning model comprises a large language model;
• receiving, from a computing device, a second user reported compatibility corresponding to the category of vehicle and the item to provide as input to the machine learning model;
• automatically generating, in response to receiving the second user reported compatibility and as output from the machine learning model, an update to the list of recommended compatibilities based on detecting the compatibility mismatch between the category of vehicle and the item;
• filtering, based on the update to the list of recommended compatibilities, a plurality of search results to obtain one or more filtered search results;
• and
• transmitting, to the computing device, additional instructions to cause the computing device to output a user interface comprising the one or more filtered search results.

A computer-implemented method performed by a first device, the computer-implemented method comprising:

• obtaining compatibility data between a plurality of items, the compatibility data comprising a list of recommended compatibilities between the plurality of items and a first user reported compatibility associated with at least one item of the plurality of items;
• training, by an artificial intelligence model manager and using at least a portion of the compatibility data, a machine learning model to detect a compatibility mismatch between a first item of the plurality of items and a second item of the plurality of items, wherein the machine learning model comprises a large language model;
• obtaining a second user reported compatibility associated with the at least one item to provide as input to the machine learning model;
• automatically generating, in response to obtaining the second user reported compatibility and as output from the machine learning model, an update to the list of recommended compatibilities based on detecting the compatibility mismatch between the first item and the second item;
• filtering, based on the update to the list of recommended compatibilities, a plurality of search results to obtain one or more filtered search results;
• and
• transmitting, to a computing second device, instructions to cause the computing second device to output a user interface comprising the one or more filtered search results.

A computer-implemented method performed by a first device, the computer-implemented method comprising:

• obtaining compatibility data between a plurality of items, the compatibility data comprising a list of recommended compatibilities between the plurality of items and a first user reported compatibility associated with at least one item of the plurality of items;
• training, by an artificial intelligence model manager and using at least a portion of the compatibility data, a machine learning model to detect a compatibility mismatch between a first item of the plurality of items and a second item of the plurality of items, wherein the machine learning model comprises a large language model;
• obtaining a second user reported compatibility associated with the at least one item to provide as input to the machine learning model;
• automatically generating, in response to obtaining the second user reported compatibility and as output from the machine learning model, an update to the list of recommended compatibilities based on detecting the compatibility mismatch between the first item and the second item;
• filtering, based on the update to the list of recommended compatibilities, a plurality of search results to obtain one or more filtered search results;
• and
• transmitting, to a second device, instructions to cause the second device to output a user interface comprising the one or more filtered search results.