Artificial intelligence-based political messaging system, application and method

A computer-implemented system for generating political messages for social media, the system interfaced with a Knowledge Base, the system comprising:

• a user interface receiving one or more inputs like a political candidate's name and a “for-or-against” selection indicating support or opposition of that candidate;
• at least one pre-trained large language model (LLM) configured to receive the one or more inputs, the large-language model configured with in-context learning using examples from data in the Knowledge Base having data categories including both political issue-oriented and candidate-oriented resources, such as local and global news, local/state/federal government statistics, public census/voter opinion polls, social media messages, comment sections of news and social media platforms about the issue, and each candidate's position/stance on and related to the issue, as disclosed by the candidate in candidate's campaign website, in candidate's own words, speeches, press releases and previous legislative voting records by candidate regarding the issue;
• a pre-trained embedding model for embedding generation, wherein the pre-trained embedding model is configured to select semantically similar parts of the candidate's profile or statistics and provide contextual information related to the candidate to the at least one pre-trained LLM;
• wherein the LLM is further configured to create messages based on user-chosen message topics/issues, candidate names, and message modifiers to personalize the user-created draft messages with subsequent draft message modifiers as needed by the user to create a personalized final draft message approved by the user incorporating the message issue, themes, tones, and other input by and to the personal satisfaction of the user;
• and
• an AI Message Generator combining information from the large language model, using unsupervised and supervised learning techniques, to create political messages for distribution by the user via email, SMS/text and/or social media messaging platforms, wherein the LLM is configured to carry out a plurality of customized tasks, including information filtering, profile creation, and knowledge extraction from the Knowledge Base, based on one or more predefined custom-made examples,
• wherein the AI Message Generator further comprises a Message Format output to include Text-to-Audio Narrated Video Message Format for which a number of visual images needed for the Text-to-Audio Narrated Video Message Format is automatically selected and determined based on the length of the original Text-Based Message's narration and an average image screen duration during playback, wherein the number of images required to accompany the length of the narration are automatically selected based on at least two of the following sources of images for the user to re-edit:
• Priority use of the Statistical Images, graphs, charts, and tables from the Knowledge Base used in the creation of the Text-based Message;
• AI/Machine Learning Interpretation of a meaning of the sentences in the Message;
• and
• Image Search Engine search results using the Message's Issue and the Message's Words/Phrases as the Search Terms for the image searches.