Technical Components

  • Machine Learning Models

    • Models like BERT or GPT are fine-tuned with a dataset of woke and non-woke posts for high accuracy.

    • Training includes examples of posts categorized by human reviewers to teach the model how to detect nuanced ideological content.

  • Database and Feedback Loop

    • A database stores flagged posts for continuous improvement.

    • Users can provide feedback on flagged posts to refine the AI’s accuracy.

  • Integration with Social Media APIs

    • The AI is integrated with Twitter’s API for real-time monitoring and automated retweets.

    • Retweet actions are programmed to include the "Woke Mark" consistently for brand recognition.

PreviousHow the AI Works

Last updated