Course Description
An interdisciplinary exploration of conversational AI through humanistic and social-scientific lenses. Examination of the technologies behind contemporary conversational interfaces widely deployed across industries and evaluation of their effectiveness, human-likeness, and social implications using frameworks from various intellectual traditions, such as intercultural pragmatics, conversation analysis, and the psychology of joint action.
Athena Title
Demyst Conversational AI
Grading System
A - F (Traditional)
Student Learning Outcomes
- Students will be able to describe the technologies behind different types of conversational AI on a nontechnical level.
- Students will be able to design simple conversational flows using accessible, no-code systems.
- Students will be able to identify major skills required for effective conversational communication (e.g. turn-taking, repair, common ground management) in both human-human and human-machine interactions.
- Students will be able to evaluate the extent to which conversational AI systems communicate in human-like ways by analyzing conversational transcripts through humanistic and social-scientific lenses.
- Students will be able to identify the causes of interactional trouble in human-machine conversations and how they were or were not resolved.
- Students will be able to evaluate social and ethical implications of the training and deployment of conversational AI systems.
Topical Outline
- I. Introduction to conversational AI: history and classification of conversational AI; properties of human conversation
- II. Processes involved in conversational AI: Automatic speech recognition; natural language understanding; dialogue management; natural language generation; text-to-speech synthesis
- III. Technologies behind conversational AI: Rule-based models; retrieval-based models; end-to-end neural models; large language models; reinforcement learning from human feedback; retrieval-augmented generation
- IV. Traditional evaluation of conversational AI: evaluation metrics; benchmarks; user ratings
- V. Understanding language in context: Referential forms and entity extraction; entity linking and coreference resolution; dialogue acts and intent classification; ellipsis resolution
- VI. Managing the local structure of interactions: turn-taking and duplex communication; conversational troubles and repair; grounding and the backchannel; discourse markers and conjunctions
- VII. Managing the global structure of interactions: Opening and closing conversations; dialogue state tracking and the common ground; sequence expansion and closing; topic management
- VIII. Personifying conversational AI: stance, indexicality and persona; epistemic stance and AI sycophancy; affective stance, emotion, and sentiment analysis; anthropomorphism and its risks
- IX. Cultural variation and conversational AI: Linguistic variation and language models; linguacultures; formality and register; indirectness and politeness
- X. Nonverbal communication, embodiment and social robots: prosody in speech technology; gaze and head movement; gesture recognition; persona effect
- XI. Social impacts and ethics: representational balance and fairness; harmlessness, honesty, and helpfulness; intellectual property issues; environmental costs; guardrails
