KO
Archive / Lecture
Research methods / Graduate research / Reading papers / AI and learning

Professor, I am still working on my topic — How many months have you been saying that?

Why can you not explain fifteen pages drafted with AI? A 62-slide lecture following fictional graduate student Minseo through topic refinement, paper reading, theory, hypotheses, methods, and advisor meetings.

Graduate students who draft with AI but find it hard to explain their research questions and methods · Hallym University Media School, graduate program · Communication Research Methods II · October 20, 2026 · 2026.10.20

Open presentation ↗ Read on the web
Fictional graduate student Minseo trying to find a topic as calendar pages pass

What this project covers

  • Narrow a broad interest into a research question with a concrete comparison.
  • Distinguish what you checked in a paper from what remains unknown.
  • Connect theory, hypotheses, measurement, and design to formulate questions for guidance.

Presentation outline

Expand the full outline
  1. Read the complete lecture
  2. Professor, I am still working on my topic
  3. Last week, you said two out of seven
  4. You already use AI
  5. We begin with that two
  6. Is a research topic a wild boar you catch with bare hands?
  7. You keep hunting for a topic without ever seeing it
  8. When does “a little more preparation” end?
  9. Why the student postpones
  10. What an instructor can examine with you
  11. Move from the sticking point to the next task
  12. State your interest
  13. Check the original
  14. Revise your sentence
  15. Bring a question
  16. Follow Minseo, a fictional master's student
  17. Plenty of interests; stuck turning them into research
  18. The file gets longer; confidence gets lower
  19. Afraid of arriving empty-handed
  20. The tabs multiply
  21. AI fills the gaps
  22. Judgment is still required
  23. I understand why Minseo keeps using AI
  24. The deadline will not wait
  25. So we make the next decision smaller
  26. I am going to Seoul. Where in Seoul?
  27. “AI and fake news” is still a broad address
  28. A concrete situation comes before academic terminology
  29. What Minseo actually wants to know
  30. Do not make it difficult already
  31. Written by AI does not mean false
  32. Start by asking AI to turn your words into search terms
  33. A request for AI
  34. What I do next
  35. You can ask before reading ten papers
  36. What you can show now
  37. An unsuccessful search is also material
  38. “Professor, they have already done everything I wanted to do!”
  39. We found an actual paper
  40. Do not jump to either conclusion yet
  41. The same six-point range can ask different things
  42. Do you share everything you consider accurate?
  43. Separate the judgments hidden inside “believe”
  44. Reading changed my question
  45. Before reading
  46. After reading the methods
  47. But you understand it immediately
  48. You do not need to understand everything before asking
  49. When you open the source, is that statement there?
  50. Find the location
  51. Open it yourself
  52. Compare the explanation
  53. Record the boundary
  54. Read the relevant parts to the necessary depth
  55. Before waiting for AI’s eleventh explanation
  56. If you keep getting stuck in the same place
  57. Bring the location you do not understand
  58. AI says my new idea is novel
  59. This time it really felt like my idea
  60. Read Study 2 before claiming a distinctive contribution
  61. Today was not a failure to find a topic
  62. If you stop at “It has already been done”
  63. Turn it into a question for the next meeting
  64. Another pile of summaries in Obsidian?
  65. Leave sentences your future self can recognize
  66. A record Minseo left after reading
  67. What you do not know also helps determine the next task
  68. This form is not a ticket to the meeting
  69. If you wait to fill every blank
  70. Record your current state as it is
  71. Does inserting a theory name create a theoretical background?
  72. Set aside the names and years for a moment
  73. Look for the reason between a label and a judgment
  74. See the label
  75. Infer whether it was reviewed
  76. Judge accuracy
  77. Find relevant studies
  78. Find your explanation before naming a theory
  79. How to ask AI
  80. What I need to check
  81. The same lunchbox, different inspection labels?
  82. An analogy helps understanding; research evidence needs a separate check
  83. I asked for a hypothesis and got eight arrows
  84. If it looks like a paper, is it my research?
  85. Explain one arrow all the way through
  86. Two AI questions are not necessarily the same hypothesis
  87. You expect a higher score, but cannot say why?
  88. If only the sentence ending becomes more confident
  89. If the direction is still hard to specify
  90. “More use means less confidence”?
  91. What I observed in class
  92. What we cannot claim without data
  93. Even with a relationship, causation remains unresolved
  94. Suppose this time we have fictional data
  95. The same data allow other explanations
  96. Should we confiscate coffee when people are tired?
  97. If both the news and the label differ, which caused the difference?
  98. Does random assignment finish the preparation?
  99. Experts keep judging as they write
  100. When someone who knows the field uses AI
  101. When a beginner assigns the same task
  102. Two AIs agree. Is the checking finished?
  103. Agreement and verification are different
  104. Assign work at a size you can check
  105. What did this study establish?
  106. But fifteen pages are due next week
  107. Understanding does not change the deadline
  108. Start by bringing out the submission requirements
  109. Put verified sentences before the draft
  110. Write down the basis you will give AI
  111. If you assign everything at once
  112. If you start with what you checked
  113. Do not turn “I do not know yet” into certainty
  114. Get one paragraph
  115. Compare it with the original
  116. Revise the scope
  117. Move to the next paragraph
  118. Do not put the whole world in your opening paragraph
  119. Plausible but unchecked
  120. A sentence you can start from now
  121. Does deleting the “unchecked” marker solve anything?
  122. In a working draft
  123. Before submission
  124. You brought this as a question?
  125. I do not reduce that burden to willpower
  126. Still, look for what you can do now
  127. Show something your instructor can help revise
  128. “What should I do?”
  129. “This is what I have checked”
  130. If told to read more, ask what to look for
  131. If you return without a direction
  132. You can ask on the spot
  133. If meeting is difficult, clarify how to communicate
  134. What you can agree on
  135. While waiting for a response
  136. Did the meeting leave only “I got scolded”?
  137. “Redo it.” What do you do after that?
  138. Separate the professor's words, your interpretation, and the next action
  139. Write a reading task with a visible endpoint
  140. Choose where to read
  141. Choose what to compare
  142. Record how your wording changed
  143. Bring it to the next conversation
  144. If you understood, where is your own sentence?
  145. Your field of interest can be different
  146. Finding one confusion is progress too
  147. Today, show the current state of your work
  148. Show where you read and where you got stuck
  149. “I have checked this far”
  150. “I find this part difficult to judge”
  151. Continue after class through questions
  152. Start the first conversation with this sentence