Syllabus

Course Description

Official OU Course catalog description: Ethics of developing Artificial Intelligence (AI) for Earth and environmental sciences (ES). Topics will include responsible conduct of research, ethical scientific conduct, ownership of ideas, algorithms and data, and ethics of developing AI for ES applications. Learning activities include active discussions and debates, writing, and projects.

Course Motivation

As our climate is changing, our risks and hazards we face are also evolving.  Understanding our changing environment is key to improving humanity and all of life on Earth’s resilience.  AI provides potential new ways to understand and predict these changes but AI is not a risk-free tool. Our goal with this goal is to learn how to create responsible and ethical AI for the Earth Sciences.

General Information

  • Class time: Tues/Thurs 1:30-2:45 NWC 5600
  • Prerequisites: CS 4013/5013, CS 4033/5033, or CS 5043 (or permission of instructor).
  • Instructor: Dr. Amy McGovern
    • Office: NWC 2506
    • Office hours: Office hours will be posted on slack.

Course Goals

The AI and ethics content will include the following, especially as they relate to Earth Sciences:

  • Bias: Do the population biases in our data create forecasting biases in the AI algorithms? Could this bias affect specific populations unfairly? How can we account for this bias in our data collection and in our learning algorithms?
  • Transparency: As we develop trustworthy AI, are the algorithms transparent to all populations? Our focus group is expert scientists, but does this create a lack of trust with the general public? What can we do to enhance the overall transparency of the algorithms for everyone?
  • Liability: Who is responsible if an AI algorithm produces an incorrect forecast and people are killed or property is lost? Are there policies that can help reduce liability? Who is responsible if a self-driving car makes a mistake?
  • Humanity: How do machines affect our behavior and interaction? How do forecasters and the general population react to AI-generated forecasts? How do we properly study trustworthy AI with human subjects? Does the difference between experts and the general public matter for the overall trustworthiness of AI?
  • Security: How do we protect the critical algorithms from deliberate errors in data? How can we protect the AI from security breaches?
  • Employment: As AI algorithms become integrated into the forecasting and scientific workflow, how does that affect the employment of current and future forecasters and scientists? Do jobs disappear or do they just change and require new training? How does this affect employment across all jobs? How does this potentially restructure society?
  • Policy: What policies exist to govern the use and development of AI? Are there any policies guiding the ethical use of AI?
  • Data sharing: When you share data used to train an AI algorithm, are there ethical issues you need to consider? How well can you anonymize the data that you release? Is there harm from sharing the data?
  • Risk analysis and communication: What are the risks of the algorithm you are developing when it is deployed? How can the algorithm’s decisions best be communicated to the end-users in a way to minimize risk of harm?

Learning Objectives

By the end of the semester, you will be able to:

  1. Identify risks of AI development and deployment with a focus on Earth Science applications
  2. Identify ways to mitigate risks for AI deployment with a focus on Earth Science applications
  3. Identify sources of bias for AI algorithms, particularly as it relates to Earth Science/geoscience applications
  4. Identify ways to reduce bias in AI algorithm development
  5. Identify the potential ethical implications of data collection
  6. Communicate AI risks effectively across disciplines, in public discourse, and politically
  7. Work effectively in interdisciplinary teams

Texts

Some of these are available for free online.  Anything that is not available for free should be available at the OU Library.

  • Race After Technology by Ruha Benjamin
  • Weapons of Math Destruction by Cathy O’Neil

Selected readings from the following will also be provided.

  • Automating Inequality by Virginia Eubanks
  • Unmasking AI by Joy Buolamwini
  • Data Feminism by Catherine D’Ignazio and Lauren F. Klein
  • On Being a Scientist published by the National Academies Press
  • Scientific Integrity and Ethics in the Geosciences by Linda Gundersen
  • Recent scientific articles on AI, ethics, and geoethics 

Learning Activities, Assignments, and Assessment

What you get out of a course depends on what you put into it! You will accumulate points on a variety of activities for this course, including the following list.

  • In-class discussions and debates
  • Writing, including both creative writing and persuasive writing
  • Semester-long project exploring an AI ethics topic in depth
  • Self-assessment of your learning

Students enrolled in the 5000 level of the course will have additional requirements for each assignment beyond the students in the 4000 level.

  • The semester-long project will require a more in-depth analysis for students enrolled in the 5000 level. In addition, 5000 level students will need to propose and create a unique project, while 4000 level students can re-create a project from the literature.
  • Students in the 5000 level will be answering additional questions in their writing assignments. In addition, their critical thinking and analysis answers will be expected to be more in-depth. For example, if a student has to explain why an AI system performed poorly in a specific task, 5000 level students would be expected to have a more in-depth analysis of how the system failed and how it could be prevented in the future.

Participating in class is one of the best ways to learn so please ask questions and attend class. All assignments will be evaluated using a rubric that will be posted when each assignment is released. This will also be available on canvas.

Grades

In order to give you a fair grade at the end of the semester, grades will be assigned based on the points for each assignment.  The grade will come from:

  • Semester project: 40% (this will be broken into small components, not all one deliverable)
  • In-class discussions and debates and class participation: 30%
  • Writing component : 30%.

As needed, grade cutoffs may move below 90/80/70/60 but will never be moved above these. Thus a grade of >= 90 will always be an A, while an 87 could potentially be an A depending on any adjustments that are needed to ensure grading is always fair.

Grading Rules

  • Grade questions: To maintain fairness in grading, any questions should be brought to the person who graded it. To maintain fairness, all disagreements about grading should be brought to our attention within one week of when the item is graded.
  • Online Grade Summary: Canvas has a grade book that I will use to store all of your grades. It is your responsibility to verify that the grades on Canvas are correct. If an error is found, notify me and I will correct Canvas.
  • Due dates: You have five free “life happens” days that will give you a no-questions-asked 24 hour extension on any assignment that is not due in-class (e.g. if you are supposed to lead a discussion and you are sick, let us know in advance and we can trade with someone!). You can miss multiple in-class discussions for illness/emergencies (see grades in canvas specifying drops per category). 
  • Sick days/Emergencies: Do not panic! If you are sick, please let me know as soon as possible and focus on getting well. We will address it when you are well. If you are dealing with an emergency, check back in after your emergency is over and we will figure out how to handle any assignments that you missed. To reduce the spread of COVID, if you are feeling unwell, please remain home but contact me to get access either to a live stream or a recorded version of the class.
  • Project: Your final project deadline will be due the last week of classes. Per university policy, you may turn this project in prior to pre-finals week if you have completed the project.

Course and University Policies

The following set of rules will help keep us all on the same page all semester and help to ensure fair treatment for all students.

  • Academic Misconduct
    • Academic misconduct hurts everyone but particularly the student who does not learn the material. All work submitted for an individual grade must be the work of that single individual. Your project code and writeups must be written exclusively by you. Use of any downloaded code or code taken from a book (whether documented or undocumented) is considered academic misconduct and will be treated as such.
    • Any significant use of generative AI to create your writing for any assignment or project will be considered a form of academic misconduct.
      • Loophole: We will do some assignments that FOCUS on ChatGPT and similar technology.  These assignments will REQUIRE you to use chatGPT and that will be specified and it is NOT misconduct to do so for these ones only!
      • Acknowledgment: I used ChatGPT Plus for inspiration in designing some quiz questions.  If you use ChatGPT or other such tools for help, acknowledge them! 
    • The outside-world allows collaboration and so do we, but there are rules to follow to ensure that you learn the material.
      • Help cannot consist of copying code or solutions. If someone offers to help you this way, they are not helping you to learn the material!
      • For the projects, you may discuss ideas with other students but you cannot share code or specific solutions.
      • For any homework assignments, you may form study groups for help at the concept level but each homework must be in your own words and you must write your study partners’ names on your homework when you turn it in. If you do not write your study partners’ names on your homework and they are similar, we will charge you with academic misconduct.
      • Do not show another student a copy of your projects or homework before the submission deadline. The penalties for permitting your work to be copied are the same as the penalties for copying someone else’s work.
      • Make sure that your computer account is properly protected. Use a good password, and do not give your friends access to your account or your computer system. Do not leave printouts or mobile drives where others might access them.
      • Upon the first documented occurrence of academic misconduct, I will report it to the Office of Academic Integrity. The Students Guide to Academic Integrity is available here.
  • University Copyright: Sessions of this course may be recorded or live-streamed. These recordings are the intellectual property of the individual faculty member and may not be shared or reproduced without the explicit, written consent of the faculty member. In addition, privacy rights of others such as students, guest lecturers, and providers of copyrighted material displayed in the recording may be of concern. Students may not share any course recordings with individuals not enrolled in the class or upload them to any other online environment. 
  • Classroom Conduct: Your classroom conduct is expected to follow the online classroom code of conduct
  • Attendance: Per CS Department and OU policy, you can be un-enrolled from the class if you do not attend the first week of class.
  • Canvas: Login to Canvas using your 4+4, using your standard OU password. If you have difficulty logging in, call 325-HELP.
  • Class discussions: We will be using slack for class discussion and help. See the join link in canvas and join during week 1.
    • Matters of personal interest should be directed to either email or a DM in slack rather than to any broad channel, e.g. informing me of an extended personal illness.
  • Religious Holidays: It is the policy of the University to excuse the absences of students that result from religious observances and to provide without penalty for the rescheduling of examinations and additional required classwork that may fall on religious holidays.
  • Incompletes: The grade of I is intended for the rare circumstance when a student who has been successful in a class has an unexpected event occur shortly before the end of the class. OU will not consider giving a student a grade of I unless the following three conditions have been met.
    1. It is within two weeks of the end of the semester.
    2. The student has a grade of C or better in the class.
    3. The reason that the student cannot complete the class is properly documented and compelling.
  • Accommodation of Disabilities: The University of Oklahoma is committed to providing reasonable accommodation for all students with disabilities. Students with disabilities who require accommodations in this course are requested to speak with the professor as early in the semester as possible. Students requiring academic accommodation should contact the Disability Resource Center for assistance at (405) 325-3852 or TDD: (405) 325-4173. For more information please see the Accessibility and Disability Resource Center. Any student in this course who has a disability that may prevent him or her from fully demonstrating his or her abilities should contact me personally as soon as possible so we can discuss accommodations necessary to ensure full participation and facilitate your educational opportunities.
  • Adjustments for Pregnancy/Childbirth Related Issues: Should you need modifications or adjustments to your course requirements because of documented pregnancy-related or childbirth-related issues, please contact your professor or the Disability Resource Center at 405/325-3852 as soon as possible. Also, see this FAQ for answers to commonly asked questions.
  • Title IX Resources: For any concerns regarding gender-based discrimination, sexual harassment, sexual assault, dating/domestic violence, or stalking, the University offers a variety of resources. To learn more or to report an incident, please contact the Sexual Misconduct Office at 405/325-2215 (8 to 5, M-F) or smo@ou.edu. Incidents can also be reported confidentially to OU Advocates at 405/615-0013 (phones are answered 24 hours a day, 7 days a week). Also, please be advised that a professor/GA/TA is required to report instances of sexual harassment, sexual assault, or discrimination to the Sexual Misconduct Office. Inquiries regarding non-discrimination policies may be directed to the EEO office.
  • Mental Health Support Services: If you are experiencing any mental health issues that are impacting your academic performance, counseling is available at the University Counseling Center (UCC). The Center is located on the second floor of the Goddard Health Center, at 620 Elm Rm. 201, Norman, OK 73019. To schedule an appointment call (405) 325-2911. For more information, please visit University Counseling Center.
  • Final Exam Preparation Period: Pre-finals week will be defined as the seven calendar days before the first day of finals. Faculty may cover new course material throughout this week. For specific provisions of the policy please refer to OU’s Final Exam Preparation Period policy.
  • Emergency Protocols: Severe Weather: If you receive an OU Alert to seek refuge or hear a tornado siren that signals severe weather
    • LOOK for severe weather refuge location maps located inside most OU buildings near the entrances
    • SEEK refuge inside a building. Do not leave one building to seek shelter in another building that you deem safer. If outside, get into the nearest building.
    • GO to the building’s severe weather refuge location. If you do not know where that is, go to the lowest level possible and seek refuge in an innermost room. Avoid outside doors and windows.
    • GET IN, GET DOWN, COVER UP.
    • WAIT for official notice to resume normal activities. 
    • Link to Severe Weather Refuge Areas
    • Severe Weather Preparedness – Video
  • Emergency Protocols: Armed Subject/Campus Intruder: If you receive an OU Alert to shelter-in-place due to an active shooter or armed intruder situation or you hear what you perceive to be gunshots:
    1. GET OUT: If you believe you can get out of the area WITHOUT encountering the armed individual, move quickly towards the nearest building exit, move away from the building, and call 911.
    2. HIDE OUT: If you cannot flee, move to an area that can be locked or barricaded, turn off lights, silence devices, spread out, and formulate a plan of attack if the shooter enters the room.
    3. TAKE OUT: As a last resort fight to defend yourself.
    4. For more information, visit OU Emergency Preparedness and Shots Fired on Campus Procedure – Video
  • Emergency Protocols: Fire Alarm/General Emergency: If you receive an OU Alert that there is danger inside or near the building, or the fire alarm inside the building activates:
    1. LEAVE the building. Do not use the elevators.
    2. KNOW at least two building exits
    3. ASSIST those that may need help
    4. PROCEED to the emergency assembly area
    5. ONCE safely outside, NOTIFY first responders of anyone that may still be inside building due to mobility issues.
    6. WAIT for official notice before attempting to re-enter the building.
    7. OU Fire Safety on Campus – Video

MODULE MENU