Fall 2023 Course Topic

Artificial Intelligence, Humans and Society

In the fall 2023 semester, IST 197G will take a look into the current state of artificial intelligence (AI), how humans impact this technology, and how this technology impacts society by asking questions like those below. If these questions pique your interest, then the fall 2023 offering of IST 197G may be perfect for you.

• What is intelligence? Can we “control” the intelligence of AI so that it is human-compatible? If so, how?
• How does AI/ChatGPT generate conversations? Are they reliable? Does this capability have risk for the society? How can we manage the risk?
• What causes an AI system (e.g., facial recognition, hiring screening) to be biased? Can such risk be reduced?
• Where does the intelligence in AI come from? Human, data, or both?
• What impacts have AI brought to healthcare? Can this be achieved without violating patient privacy?
• What are ethics-related issues for AI?
• Can AI explain itself so that we (humans) can better calibrate our trust on AI?

Using real-world AI applications, you will be able to connect the history of AI to its advancement in the 21st century, as well as understand the benefits and potential risks of AI. This course is delivered with significant student and instructor interaction with computers and digital media. This course has no pre-requisites, does not involve programming, and satisfies General Education requirements in the knowledge domain of Social and Behavioral Sciences (GS).

The course consists of four modules outlined below.

1. Intelligence, ChatGPT, and human-compatible AI: The first module starts with two fundamental questions: “What is intelligence?”, “Whether AI should be controlled?”. Using ChatGPT as an example, students will engage in discussions on whether the intelligence of AI should be “controlled”? If so, what does it mean? If not, what risks we may face in the future.
2. Bias and Societal Risk of AI: This module uses real-world cases to expose some of the bias of AI systems (e.g., facial recognition, hiring screening) and the risk they can introduce to the society. Using AI applications in social media and law enforcement, students will be asked to reflect on ethical issues of AI design strategies adopted by Big Techs.
3. Self-driving Cars and Precision Health: The third module introduces two real-world AI applications: self-driving cars and AI application in medicine. In both areas, students will learn efforts and challenges in the early period of AI in each of these two areas, as well as recent acceleration. Ethical issues and exemplar data harvesting approaches that protect privacy of patients will be discussed.
4. Human Perception and Trust on AI: The final module introduces students to human perception and trust calibration on AI. The module will also invite students to reflect on the difference of explainable AI versus black-box AI on human-AI trust and its potential impact on the society.

Special topics courses are offered infrequently to explore, in depth, a comparatively narrow subject that may be topical or of special interest.