Conversational and Multimedia Deception

This unit examines how advanced conversational AI, voice synthesis, deepfake imagery, and robotics can be composed and orchestrated to create persuasive or deceptive communications. Students will study technical mechanisms (LLMs, prompting, Monte Carlo simulation, relays), analyze risks and unintended failures, and prepare defensive, ethical, and policy responses in controlled, evaluative settings.

Learning Objectives

• Analyze conversational AI strategies including context crafting, prompting, relays, and bot orchestration to identify technical mechanisms that enable persuasive or deceptive interactions
• Apply simulation and modeling techniques, including Monte Carlo methods, to quantify and compare the effectiveness and variability of automated social engineering scenarios in controlled experiments
• Demonstrate the creation and forensic analysis of multimodal synthetic artifacts (text, voice, and image deepfakes) to evaluate technical fingerprints, limitations, and detection approaches
• Evaluate the ethical, legal, cyber‑security, and societal implications of AI-enabled deception and robotics, and propose mitigation, detection, and policy-based strategies for reducing harm