This course is about
Sophisticated computer programs and robots already have significant ethical impacts on our lives. Consider: automated computer systems that decide which social-media posts are shown to us, or that decide which loan applications are denied, or that decide which job applications are forwarded, or that recommend which incarcerated people are granted parole. Consider self-driving cars that must decide whether to favor the car’s occupants or outside pedestrians. Consider autonomous robotic health-care assistants that cajole patients to do things they might not want to do. Consider autonomous military drones that decide whom to kill. How do we check that those systems are being fair and unbiased, and treating people ethically or at least legally? Crucially, how can we possibly program ethics or human morality into computer programs and robots?
This course emphasizes the importance of applying human moral psychology to these issues. Every topic relies on human psychology: What is the psychology of fairness? What is the psychology of explaining or justifying an action? What is the psychology of responsibility and blame? What is the psychology of attributing moral standing or rights? Do those judgments vary across cultures? Importantly, what are the various psychological and evolutionary functions of human morality, and should those same functions be mimicked or acknowledged in artificial moral agents?
This course is structured as readings with discussion. Students are expected to do extensive reading every week, and to be prepared to discuss the readings in class.
Please purchase this book:
All required articles will be posted in Canvas.
The schedule below is the plan as of Jan 04, 2022. The exact dates or readings may change as the semester progresses. Any changes will be announced in class and on Canvas.
| Week | Day | Theme | What |
|---|---|---|---|
| 1 | Tues | Overview | Welcome and course overview. (In lecture but not assigned reading: Scheutz & Malle (2018): Moral robots; and, Liao (2020): A short introduction to the ethics of artificial intelligence. FYI: Video featuring Scheutz and Malle.) |
| 1 | Thur | Quiz. Christian (2020): The alignment problem: machine learning and human values. Prologue & Intro & Ch 1. FYI: Brian Christian summarizes the book in this video, and has a partial summary with extended discussion in this video. |
|
| 2 | Tues | Quiz. Christian (2020): The alignment problem: machine learning and human values. Ch 2. (FYI, tangential: Using social media for policing.) |
|
| 2 | Thur | Quiz. Christian (2020): The alignment problem: machine learning and human values. Ch 3. | |
| 3 | Tues | Quiz. Christian (2020): The alignment problem: machine learning and human values. Ch’s 4 & 5. | |
| 3 | Thur | Quiz. Christian (2020): The alignment problem: machine learning and human values. Ch’s 6 & 7. | |
| 4 | Tues | Quiz. Christian (2020): The alignment problem: machine learning and human values. Ch’s 8 & 9 & Conclusion. | |
| 4 | Thur | Exam. | |
| 5 | Tues | Crowdsourcing morality | In-class interactive: Decide for whom the car tolls; and, Ask Delphi |
| 5 | Thur | Quiz. Awad et al. (2018): The moral machine experiment; and, Bigman & Gray (2020): Life and death decisions of autonomous vehicles; and, Awad et al. (2020): Reply to: life and death decisions of autonomous vehicles. Try it yourself: Decide for whom the car tolls. FYI: Rahwan’s TED talk. (In lecture but not assigned reading: Gill (2020): Blame it on the self-driving car: how autonomous vehicles can alter consumer morality; and, Novak (2020): A generalized framework for moral dilemmas involving autonomous vehicles: a commentary on Gill.) |
|
| 6 | Tues | Quiz. Jiang et al. (2021a): Delphi: towards machine ethics and norms; and, Vincent (2021): The AI oracle of Delphi uses the problems of Reddit to offer dubious moral advice; and, Jiang et al. (2021b): Towards machine ethics and norms. Try it yourself: Ask Delphi FYI: NYT blog. (In lecture but not assigned reading: Talat et al. (2021): A word on machine ethics: a response to Jiang et al.(2021).) |
|
| 6 | Thur | Implementation | Quiz. Malle & Scheutz (2014): Moral competence in social robots. |
| 7 | Tues | Quiz. First parts of Cervantes et al. (2020): Artificial moral agents: a survey of the current status. | |
| 7 | Thur | Quiz. Remaining parts of Cervantes et al. (2020): Artificial moral agents: a survey of the current status. (In lecture but not assigned reading: Tolmeijer et al. (2020): Implementations in machine ethics: a survey.) |
|
| 8 | Tues | Quiz. Zoshak & Dew (2021): Beyond Kant and Bentham: how ethical theories are being used in artificial moral agents. FYI: Video by Zoshak summarizing the article. |
|
| 8 | Thur | Quiz. Holyoak & Powell (2016): Deontological coherence: a framework for commonsense moral reasoning. (Especially the section on “Deontological coherence as constraint satisfaction.”) | |
| 9 | Tues | Review. | |
| 9 | Thur | Exam. | |
| Break | |||
| 10 | Tues | Explainability | Quiz. Gunning & Aha (2019): DARPA’s explainable artificial intelligence (XAI) program. |
| 10 | Thur | Quiz. Yong et al. (2020): Not so much rational but rationalizing: humans evolved as coherence-seeking, fiction-making animals. (In lecture but not assigned reading: Miller (2019): Explanation in artificial intelligence: insights from the social sciences. Also preview “moral dumbfounding.”) |
|
| 11 | Tues | Corruptor or Advisor | Quiz. Köbis et al. (2021): Bad machines corrupt good morals. (In lecture but not assigned reading: Zhu et al. (2020): Blame-laden moral rebukes and the morally competent robot: a Confucian ethical perspective.) FYI: robots in religion. |
| 11 | Thur | Responsible Use | Quiz. Rahwan (2018): Society-in-the-loop: programming the algorithmic social contract. (In lecture but not assigned reading: Hagendorff (2020): The ethics of AI ethics: an evaluation of guidelines.) FYI: Cf. the Belmont principles for human subjects. |
| 12 | Tues | Attitudes toward robots | Quiz. Chapter 3 “The Machine” of Wegner & Gray (2016): The mind club: who thinks, what feels, and why it matters. (In lecture but not assigned reading: Gamez et al. (2020): Artificial virtue: the machine question and perceptions of moral character in artificial moral agents.) |
| 12 | Thur | Perspectives | Quiz. Martinho et al. (2021): Perspectives about artificial moral agents. |
| 13 | Tues | Human moral psychology | Quiz. Pinker (2008): The moral instinct. And, begin discussion of Haidt & Kesebir (2010): Morality. |
| 13 | Thur | Quiz. Haidt & Kesebir (2010): Morality, continued. | |
| 14 | Tues | Human morality involves relationships | Quiz. Rai & Fiske (2011): Moral psychology is relationship regulation: moral motives for unity, hierarchy, equality, and proportionality. (In lecture but not assigned reading: Tomasello (2016): A natural history of human morality.) See also Professor Kruschke’s course in moral psychology. |
| 14 | Thur | Machine understanding of relationships | Quiz. Tentatively: Teng et al. (2021): Toward jointly understanding social relationships and characters from videos; and, Xu et al. (2021): Socializing the videos: a multimodal approach for social relation recognition. |
| 15 | Tues | Machine ethics vs human morality | Prof. Kruschke’s perspective: • Machine obligations vs human obligations (to self, to others in various roles, to various groups, to principles). • Control of machines vs control of humans (through social norms and social sanctions such as reputation, gossip, ostracism). • Machine explanation vs human explanation, justification, recruitment of allies, persuasion. • Machine motivation vs evolved human emotion, cognition, and motivation for social functions. |
| 15 | Thur | Review and Overview (or possibly study day) | |
| Finals | Final Exam, see registrar info for date and time |
The items below do not exhaust everything to learn from this course, but here are at least a few key points.
Most classes begin with a short quiz about that day’s reading. The purpose of the quiz is to motivate you to prepare for in-class discussion and to recognize your preparation. Each day’s quiz consists of a small number of short-answer or multiple choice questions relevant to that day’s reading. The quiz starts promptly at the beginning of class. Each quiz is worth up to 10 points.
Your lowest 4 quiz scores will be dropped. Thus, if you must miss a day because of illness or personal reasons, that day simply counts as a zero and will be one of your 4 dropped scores. There are no make-up quizzes for missed days. The only way that a make-up for a missed quiz could be considered is if you had to miss more than 4 quiz days for reasons beyond your control, in which case notify Prof. Kruschke as soon as possible.
There are 3 exams. The purpose of the exams is to help you retrospectively synthesize the readings. Each exam will focus mostly on material from the immediately preceding part of the course, but there will be some questions that comprehensively integrate across all previous material. Each exam is worth 100 points, for 300 points total. All exams are mandatory. Make-up exams are given only if you had to miss an exam for reasons beyond your control, in which case you must notify Prof. Kruschke immediately and in advance if possible.
Graduate students enrolled in P657, but not undergraduates in P457, will compose a research proposal based on the course readings. The proposal should include an introduction that explains how the course readings motivate the proposed research and what novel contribution would be provided by the research. The proposal should explain the methods and procedures of the proposed research, and how they relate to the course readings. The proposal should also explain anticipated results and what type of statistical procedures might be used. The proposal does not need to include simulated (or actual) data or statistical analyzes. The proposal should not exceed 5,000 words. The proposal will count as 15% of the final course grade.
Letter grades in P457 are based on your total points, as a percentile relative to other students in the course. Percentiles are established for P457 separately from P657. There are no pre-set point cutoffs for specific letter grades, nor is there a pre-set quota for how many students can receive A’s or B’s, etc. Past experience suggests that there will be approximately 30% A’s, 40% B’s, 25% C’s, and 5% D’s. For P657, grades will be assigned as is typical for a graduate readings course.
Our online hub for the course is IU Canvas (https://canvas.iu.edu/). If you do not have access to Canvas, please notify Prof. Kruschke immediately. All announcements and discussion will be consolidated into the P457 Canvas site, therefore students who are enrolled in P657 will be given access to the P457 Canvas site.
Prof. John Kruschke, kruschke@indiana.edu. Office hours by appointment; please do ask. Meetings may be held via Zoom.
All information in this document is subject to change. Changes will be announced in class.
As a student at IU, you are expected to adhere to the standards detailed in the Code of Student Rights, Responsibilities, and Conduct (Code). Academic misconduct is defined as any activity that tends to undermine the academic integrity of the institution. Violations include: cheating, fabrication, plagiarism, interference, violation of course rules, and facilitating academic dishonesty. When you submit an assignment with your name on it, you are signifying that the work contained therein is yours, unless otherwise cited or referenced. Any ideas or materials taken from another source for either written or oral use must be fully acknowledged. All suspected violations of the Code will be reported to the Dean of Students and handled according to University policies. Sanctions for academic misconduct may include a failing grade on the assignment, reduction in your final course grade, and a failing grade in the course, among other possibilities. If you are unsure about the expectations for completing an assignment or taking a test or exam, be sure to seek clarification from your instructor in advance.
Several commercial services have approached students regarding selling class notes/study guides to their classmates. Selling the instructor’s notes/study guides in this course is not permitted. Violations of this policy will be reported to the Dean of Students as academic misconduct (violation of course rules). Sanctions for academic misconduct may include a failing grade on the assignment for which the notes/study guides are being sold, a reduction in your final course grade, or a failing grade in the course, among other possibilities. Additionally, you should know that selling a faculty member’s notes/study guides individually or on behalf of one of these services using IU email, or via Canvas may also constitute a violation of IU information technology and IU intellectual property policies; additional consequences may result.
The faculty member teaching this course holds the exclusive right to distribute, modify, post, and reproduce course materials, including all written materials, study guides, lectures, assignments, exercises, and exams. While you are permitted to take notes on the online materials and lectures posted for this course for your personal use, you are not permitted to re-post in another forum, distribute, or reproduce content from this course without the express written permission of the faculty member. Any violation of this course rule will be reported to the appropriate university offices and officials, including to the Dean of Students as academic misconduct.