Generative AI Models and Prompting

This unit surveys major generative AI models (text, image, audio, and music) and examines how prompts shape their behavior, outputs, and practical applications. Students will explore model capabilities, limitations, ethical implications, and future trajectories to prepare for applied prompt engineering and system integration in subsequent units.

Learning Objectives

• analyze the architectures, intended uses, and prompt-handling behaviors of ChatGPT, DALL-E, CTRL, T2T, BERT, Tacotron, and MuseNet and compare their strengths and weaknesses
• apply prompt-design principles to create, test, and iteratively refine prompts across at least three different model types (text, image, audio/music) and document outcome differences
• evaluate limitations, biases, and ethical or societal implications for each model and recommend mitigation or responsible-use strategies
• demonstrate an understanding of potential future developments and practical opportunities for integrating these generative models into software and systems development