Example: Journal Club

If your learners are assigned a presentation of an article for a journal club, they could first use AI to try to answer their questions about methodology. Imagine they are assigned this article and want to learn more about Bayesian meta-analysis.

Prompt: "I am reading a paper that is a Bayesian meta-analysis. Can you explain to me what that means?"

This response goes on, but it gives a nice overview of Bayesian Meta-Analysis. You can even upload the PDF of the study directly and get a more detailed output.

Prompt: "Can you tell me how Bayesian meta-analysis was used in this study?" (s13054-024-04831-4.pdf)

Now, is it possible that learners will just ask AI to generate their journal club presentation? Yes. But this is where educators can step in and think about how they can use the AI to help with the learning process. We could examine AI's assumptions and its conclusions and evaluate the accuracy of its outputs. We could ask follow up questions and clarifications as they arise.

Inaccuracies could arise, and this is where educators need to be able to help learners critically evaluate the output. We also need to consider the alternatives if learners don't use AI. They might look at Wikipedia or another online resource to learn about Bayesian meta-analysis, which in turn might have its own inaccuracies. There is no perfect resource.