Proactively-generated content creation based on tracked performance

A system comprising:

• a memory comprising instructions;
• and
• one or more computer processors, wherein the instructions, when executed by the one or more computer processors, cause the system to perform operations comprising:
• transmitting a plurality of first items to first computer devices for presentation to users, the first items being multimodal, each first item having values for one or more attributes, the plurality of first items comprising items with different values for the one or more attributes, the one or more attributes comprising attributes associated with a brand, a product, or a user;
• tracking performance of the transmitted plurality of first items, wherein the performance comprises information on a level of interaction of users for each transmitted item;
• training a machine-learning model with training data comprising performance information of items after presenting the items to users, the machine-learning model generating relevance scores of attributes for generating items;
• utilizing the machine-learning model to compare the performance of the different values of the attributes based on the level of interaction of the users;
• selecting values of one or more attributes based on the performance of the different values of the attributes in the plurality of first items;
• generating a plurality of prompts based on the selected values of the one or more attributes, each prompt comprising text in human language that describes a desired output from a generative artificial intelligence (GAI) tool;
• providing as input each from the plurality of prompts to one or more GAI tools that generate a plurality of second items;
• providing a user interface (UI) with an option to select from the plurality of second items;
• receiving in the UI a selection of selected second items for transmittal;
• and
• transmitting the selected second items to second computer devices of one or more users.

A non-transitory machine-readable storage medium including instructions that, when executed by a machine, cause the machine to perform operations comprising:

• transmitting a plurality of first items to first computer devices for presentation to users, the first items being multimodal, each first item having values for one or more attributes, the plurality of first items comprising items with different values for the one or more attributes, the one or more attributes comprising attributes associated with a brand, a product, or a user;
• tracking performance of the transmitted plurality of first items, wherein the performance comprises information on a level of interaction of users for each transmitted item;
• training a machine-learning model with training data comprising performance information of items after presenting the items to users, the machine-learning model generating relevance scores of attributes for generating items;
• utilizing the machine-learning model to compare the performance of the different values of the attributes based on the level of interaction of the users;
• selecting values of one or more attributes based on the performance of the different values of the attributes in the plurality of first items;
• generating a plurality of prompts based on the selected values of the one or more attributes, each prompt comprising text in human language that describes a desired output from a generative artificial intelligence (GAI) tool;
• providing as input each from the plurality of prompts to one or more GAI tools that generate a plurality of second items;
• providing a user interface (UI) with an option to select from the plurality of second items;
• receiving in the UI a selection of selected second items for transmittal;
• and
• transmitting the selected second items to second computer devices of one or more users.