Reflect on and describe the impacts of misinformation and the role of regulation
Evaluate how images can change the communication and quality of textual information
Introduce the assignment, make sure students have registered to use the AI image-generator tools that will be used, and demonstrate how to use them, if needed. Students will need computers. If in-class laptops are limited, the assignment can be adapted for small groups or given as a take-home assignment.
- First, introduce the assignment: students will be illustrating a hoax, “corroborating” false narratives, and providing “evidence” that would be characterized as misinformation.
- Go to Hoaxes on Wikipedia.
- Each student reviews the list of hoaxes and chooses one to work on for the assignment. They can Google around for more context and understanding.
- Using AI image-generation tool(s), students should illustrate the hoax, as if it were a real event, place, person, or phenomenon. In other words, ask them to furnish “evidence” that would support the “truth” of the hoax. This could include AI-generated photos, archival material, other documents, or anything else the student chooses to produce. For example, if you are trying to offer proof to corroborate the hoax that the Italian soccer player Damiano Tommasi was an expert noted for “in-depth knowledge and appreciation” of the philosopher Søren Kierkegaard, you might prompt Midjourney with the following: An ultra-realistic photographic image of the Italian soccer player Damiano Tommasi sitting in the bleachers on the sidelines of a soccer pitch, deeply engrossed in reading a book, on the cover of the book the words “Fear and Trembling” are printed, natural light, low contrast, 4k
- Each student prepares a 2-3 slide presentation to give in-class that conveys the hoax as “truth.”
- Optional/time permitting: student volunteers share what they created.
- Discussion (15 min)
Here are some sample discussion questions:
- What was hardest about this assignment? What was most intriguing?
- What, if anything, surprised you?
- Do you think AI image-generation tools are being used to produce misinformation and to deceive? Is this different, in kind, from previous forms of image manipulation and propaganda?
- What would be some potential long-term (serious) harms of such technologies? Should they be regulated? If so, by whom? Should certain forms of transparency of process be required? If so, by whom?
- What are the potential long-term benefits of such technologies? Do the benefits distribute evenly or unevenly? Who stands to benefit the most? Who stands to be harmed the most?
- Take-home (and optional) written reflection: Write a one-page description of what you created and the tools you worked with, your reflections on the technology, and how you got the outputs you were looking for (or not), including the prompts you used and tried. Then discuss risks of misuse, and end with a position on how and whether you think these tools should be limited in their use by companies and journalistic outlets.
- “China Sows Disinformation About Hawaii Fires Using New Techniques,” The New York Times