Critiquing AI Output | AI for Learning @ColumbiaCTL
Critiquing AI Output | AI for Learning  @ColumbiaCTL
Uploaded September 2024 | Updated November 2026, 2 hours ago
This video offers a demonstration on how to guide students in critiquing output generated by AI. By engaging in this type of activity, students become aware of the strengths and limitations of AI, identify biases and inconsistencies in the AI-generated output, make judgments about the quality, credibility, and relevance of it, and develop metacognitive skills. Instructors can teach students how to think critically about AI-generated output, make space in class for students to practice these important skills and facilitate a conversation after the experience to share what was learned.

The prompt used in this video is provided below and can be adapted for each course context.

For more resources on leveraging AI for teaching and learning, visit the Columbia Center for Teaching and Learning’s website at ctl.columbia.edu.

Prompt: Write a five-sentence summary of the following text. Be sure to highlight the main point of the article.

Text to summarize: [insert text of your choice]

Generative AI Policy: https://provost.columbia.edu/content/office-senior-vice-provost/ai-policy

CUIT ChatGPT Enterprise: https://www.cuit.columbia.edu/content/chatgpt-enterprise
Critiquing AI Output | AI for LearningCWR 3.9 IntroductionLearning Innovation and the Future of Higher Education with Joshua Kim and Edward MaloneyFOS4 | 4.2 The Ice Core Record  | Section 4: Past ClimateDead Ideas in Grading with Jenny DavidsonLet’s Stop Relying on Biased Teaching Evaluations with Joanna WolfeFOS4 | 6.2 The Paris Accord | Section 6: Global Plans and ProspectsDead Ideas in Intercultural Development with Tara HarveyThe Secret to a Connected Classroom? Names Matter with Michelle Miller.Lance Weiler and Lydia Chilton | Faculty Voices on AI in the Classroom: Digital StorytellingTeaching Development in Doctoral Education: Let’s Ask the Grad Students!The Role of CTLs and Institutional Change with Mary Wright
ColumbiaLearn |

Critiquing AI Output | AI for Learning

SHARE TO X SHARE TO REDDIT SHARE TO FACEBOOK WALLPAPER