YouTube2Text

【AI時代の教育論】卒業論文がなくなる?レポート試験は意味がない?AIと大学の最前線(解説:京都大学准教授 柴山桂太) — Transcript

by 『表現者クライテリオン』チャンネル · 3,692 words · 662 segments · language en · Watch on YouTube

Full transcript

  1. 0:00Today, we have Professor Shibayama with
  2. 0:02us to discuss how the classroom is
  3. 0:03changing with the advent of ChatGPT.
  4. 0:05It’s a pleasure to be here.
  5. 0:06>> Thank you for joining us.
  6. 0:08>> You teach at Kyoto University, and I’d
  7. 0:11like to ask you candidly how the
  8. 0:12emergence of AI has changed the
  9. 0:14educational landscape.
  10. 0:16>> Hmm. Well, let’s see. For one thing,
  11. 0:18report-based exams have become somewhat
  12. 0:20, well...They’ve become meaningless,
  13. 0:22haven’t they?
  14. 0:23>> Yes.
  15. 0:24In my experience, while the AI boom has
  16. 0:26been going on for a few years, its
  17. 0:28performance has improved dramatically
  18. 0:31recently.
  19. 0:33In the past, there were "hallucinations
  20. 0:36," like citing non-existent literature—
  21. 0:38though some of that remains—and the
  22. 0:40writing had a very "AI-like" quality to
  23. 0:43it.
  24. 0:45>> There were times when you could tell if
  25. 0:47a report was written by AI just by
  26. 0:49looking at it, but it’s becoming harder
  27. 0:51and harder to distinguish.
  28. 0:56So, report exams at universities were
  29. 0:58originally meant to train students to
  30. 1:00review what they learned in class, think
  31. 1:03for themselves, and then construct an
  32. 1:05argument.
  33. 1:08But now, AI can research and write
  34. 1:10everything in one go.
  35. 1:12>> Yeah, it just spits it out, so it’s
  36. 1:14pointless, isn’t it? Exactly.
  37. 1:17Because of that, universities—not just
  38. 1:19mine, but across the board—are moving
  39. 1:22away from the old way of doing report
  40. 1:24exams where you just write a final paper
  41. 1:26for credit.
  42. 1:28Now, they’re asking for comments every
  43. 1:30single class, or even requiring students
  44. 1:33to write things out by hand.
  45. 1:35>> Mm-hmm.
  46. 1:35>> Or switching to traditional written
  47. 1:37exams at the end of the term.
  48. 1:40>> Some are even assigning reports and then
  49. 1:42asking students to critique what the AI
  50. 1:44produced.
  51. 1:46>> Ah.
  52. 1:47It seems some professors are moving
  53. 1:49toward exams that assume AI use,
  54. 1:50focusing on developing critical thinking
  55. 1:53by having students analyze the AI's
  56. 1:54output.
  57. 1:56>> I see.
  58. 1:56>> So, it’s fair to say that the
  59. 1:58educational field, including
  60. 2:00universities, is still groping for ways
  61. 2:02to respond to such rapid technological
  62. 2:04changes.
  63. 2:04>> Yes. But if things continue this way,
  64. 2:07well, regarding things like graduation
  65. 2:09theses, if AI keeps progressing at this
  66. 2:12current pace, it will reach a point
  67. 2:14where it can write long academic papers
  68. 2:16without any logical breakdowns.
  69. 2:18>> Yes.
  70. 2:19>> In fact, when looking at students '
  71. 2:21papers, the number of references has
  72. 2:23suddenly increased, hasn't it?
  73. 2:26>> Yeah.
  74. 2:26>> Especially English literature, too.
  75. 2:28>> Yes.
  76. 2:29>> Well, AI is strong with English. Right,
  77. 2:31if you research a topic, the papers just
  78. 2:32come pouring out. And then you just
  79. 2:34download the PDF. Don't you? And if you
  80. 2:35have the AI read the PDF, it will even
  81. 2:37summarize it for you.
  82. 2:38>> Yeah.
  83. 2:39>> So, just by sitting at home and typing
  84. 2:41away while interacting with AI,
  85. 2:43information flows in one after another;
  86. 2:46writing a paper involves citing and
  87. 2:48referencing previous research, and that
  88. 2:50becomes, how should I put it,
  89. 2:52dramatically easier to do.
  90. 2:55And having an AI write a paper for you
  91. 2:58becomes so easy as well.
  92. 2:59>> So, I'm sorry, but our...What was the
  93. 3:01original purpose of the act of writing a
  94. 3:04thesis anyway?
  95. 3:05>> Yeah. That's right. This is often
  96. 3:07discussed, but personally, the reason
  97. 3:09why universities have graduation theses
  98. 3:12is that, after all, it is...
  99. 3:14>> Something cultivated throughout
  100. 3:16elementary, middle, and high school.
  101. 3:19Well, it's the final test, if you will,
  102. 3:21of that intellectual ability, I think.
  103. 3:25Because...
  104. 3:25>> Yeah.
  105. 3:27Writing a thesis isn't just writing a
  106. 3:29book report; it’s about finding a
  107. 3:31problem yourself—for instance, noting
  108. 3:33that previous research hasn't addressed
  109. 3:35it yet, or hasn't covered it
  110. 3:37sufficiently.
  111. 3:39>> Yeah.
  112. 3:39>> For example, if there is a known area of
  113. 3:42what we already understand and an
  114. 3:44unknown area, the things humans have
  115. 3:46actually managed to clarify are really
  116. 3:48just this much of the entire universe.
  117. 3:50>> And there is still a vast amount of the
  118. 3:52unknown. In every field, even if you
  119. 3:54think you understand it, there are
  120. 3:55actually quite a few ambiguous parts.
  121. 3:57>> Yeah.
  122. 4:01So, you define your own problem, saying
  123. 4:03"this is what I'll focus on," and while
  124. 4:06building on past research, you
  125. 4:07incorporate your own findings or perhaps
  126. 4:10sources that wouldn't normally be used
  127. 4:12for that topic, to slightly expand the
  128. 4:14realm of the known.
  129. 4:20Well, for a bachelor's, master's, or
  130. 4:22doctoral degree, it's about the extent
  131. 4:24to which you've turned the unknown into
  132. 4:26the known.
  133. 4:29For a doctorate, you need to turn
  134. 4:31something much larger into the known,
  135. 4:33but even for a bachelor's thesis, you
  136. 4:35still need to show some...
  137. 4:37>> Yeah.
  138. 4:40>> It's about having them do something
  139. 4:41where it can be objectively verified
  140. 4:43that the realm of the known has expanded
  141. 4:45because they wrote that thesis.
  142. 4:48And that is...yeah.
  143. 4:49>> I personally feel that the world of
  144. 4:51industry and business is quite similar.
  145. 4:53>> In business, there's no point in just
  146. 4:55doing what everyone else is doing, is
  147. 4:56there? That's right.
  148. 4:57>> So, you need to get involved in areas
  149. 4:59where a sufficient market doesn't exist
  150. 5:01yet or things that are still unknown...
  151. 5:03>> Yeah.
  152. 5:05...
  153. 5:06>> use your intellect there to...
  154. 5:09>> And by actually taking action there...
  155. 5:11...
  156. 5:11>> expanding the market...
  157. 5:12>> Creating value-added, yes, it truly is "
  158. 5:15deep" value. Creating value-added is...
  159. 5:17...
  160. 5:17>> very much like creating islands of the
  161. 5:19known within the unknown.
  162. 5:20>> Yeah.
  163. 5:22That's why I think writing a thesis—
  164. 5:24especially in the humanities—which
  165. 5:26everyone does, is relevant because those
  166. 5:28graduates go on to become business
  167. 5:30people, and in the business world, they
  168. 5:32work hard to create value-added.
  169. 5:36I think the vector of their thinking is
  170. 5:37quite similar. And in fact, in my
  171. 5:39experience...
  172. 5:41>> I've met many practitioners, and those I
  173. 5:43consider truly great practitioners are
  174. 5:45often the type who I feel would have
  175. 5:47succeeded as scholars as well.
  176. 5:50That type isn't just naturally smart;
  177. 5:52they also research things very
  178. 5:54thoroughly.
  179. 5:54>> Yeah.
  180. 5:55>> And they discuss things with many people
  181. 5:57. There are many people of that type,
  182. 5:58aren't there? Well, not everyone, of
  183. 6:01course. Yeah.
  184. 6:06>> I’ve met many famous entrepreneurs and
  185. 6:08business people who I thought could have
  186. 6:10become incredible scholars if they had
  187. 6:11pursued an academic path.
  188. 6:17So, in that sense, I think the way they
  189. 6:20use their brains is quite similar, and
  190. 6:22while writing a thesis might seem
  191. 6:24unrelated to the real world at first
  192. 6:26glance,
  193. 6:27>> Yeah.
  194. 6:28>> the ability to research, think, discuss
  195. 6:31, and correct errors actually leads to
  196. 6:33cultivating a type of transferable
  197. 6:35intelligence.
  198. 6:36And since a graduation thesis takes a
  199. 6:38long time to prepare—six months to a
  200. 6:41year—and requires a significant
  201. 6:42investment of time,
  202. 6:44>> I personally believe it has contributed
  203. 6:46quite a bit to developing those kinds of
  204. 6:49skills.
  205. 6:50But now, with AI,
  206. 6:52>> Yes.
  207. 6:54you can write something decent without
  208. 6:56using much brainpower at all.
  209. 6:58Of course, since AI learns from past
  210. 7:01data, it doesn't write anything
  211. 7:02particularly groundbreaking, but it
  212. 7:05allows you to skip the entire essential
  213. 7:07process of thinking, discussing, and
  214. 7:09revising.
  215. 7:10>> Yeah.
  216. 7:11>> The amount of effort required is
  217. 7:12significantly reduced. It feels like it
  218. 7:14loses its meaning.
  219. 7:16That’s why some of us are saying that
  220. 7:18university graduation theses might
  221. 7:20eventually become oral exams.
  222. 7:22>> Right there on the spot.
  223. 7:24>> Exactly, on the spot.
  224. 7:26>> Instead of saying "the graduation thesis
  225. 7:28presentation begins," we’d say "the oral
  226. 7:31examination begins.""
  227. 7:33>> You wrote about this topic, right?"
  228. 7:36"So, tell us what your theme was and
  229. 7:38what kind of discoveries you made."
  230. 7:42>> By making them talk and answer questions
  231. 7:44, they have to have thought about it,
  232. 7:46researched it, and internalized it
  233. 7:47themselves to respond.
  234. 7:49I’ve heard that universities in medieval
  235. 7:51Europe were originally based on oral
  236. 7:53exams,
  237. 7:54>> Oh, really?
  238. 7:56>> so I think we might be shifting back
  239. 7:58toward that.
  240. 7:59But professor, when building a logical
  241. 8:01argument, I think you raise a question
  242. 8:04and then add your own interpretation to
  243. 8:06that foundational information, right?
  244. 8:08>> Mm-hmm.
  245. 8:09>> I think the main help AI provides is the
  246. 8:11task of gathering that foundational
  247. 8:13information. Yeah, yeah, yeah.
  248. 8:15>> Which means that...
  249. 8:17>> Well, if you're prone to relying only on
  250. 8:19your own interpretation to write that
  251. 8:21logic,
  252. 8:22>> I sort of wonder if that might actually
  253. 8:23work well for some people.
  254. 8:25>> Yeah. In what way, for example?
  255. 8:28>> Since you can spend more time broadening
  256. 8:29or deepening your own interpretation
  257. 8:31rather than researching,
  258. 8:32>> you might be able to write something
  259. 8:34more dramatic.
  260. 8:34>> Yeah. Ah. Well, that is one possibility
  261. 8:37, isn't it? If you use AI well, I
  262. 8:38suppose so.
  263. 8:39>> Yeah.
  264. 8:40>> So, if you re-examine the information AI
  265. 8:42gives you yourself, or if it feels
  266. 8:44different from your own intuition, you
  267. 8:46can have a dialogue with it there.
  268. 8:49>> I think it's possible to come up with
  269. 8:51interpretations or ideas that you never
  270. 8:53had before from that. But to do that,
  271. 8:55you need to have a certain amount of
  272. 8:57sense and knowledge accumulated on your
  273. 8:59end first.
  274. 9:00>> Yeah.
  275. 9:00>> Otherwise, you'll just be dragged along
  276. 9:02by the AI, right?
  277. 9:04>> The problem is, well, people in their
  278. 9:0630s and up today, those who have managed
  279. 9:09without AI all this time.
  280. 9:13They have that experience already, and
  281. 9:15have accumulated knowledge and
  282. 9:17background, so they're fine, but the
  283. 9:19generation coming up now is the
  284. 9:20AI-native generation.
  285. 9:23>> Yes.
  286. 9:25I was just talking to a school teacher
  287. 9:27earlier.
  288. 9:28Even at around ten years old, they're
  289. 9:30using tablets now, right? When the
  290. 9:32teacher asks if they understand, they go
  291. 9:34tap-tap-tap and just read out, "Yes, it
  292. 9:36is such-and-such," and they don't even
  293. 9:38think there's anything wrong with that,
  294. 9:41the kids.
  295. 9:41>> Ah.
  296. 9:42>> So, from the time they're children, if
  297. 9:43they don't know something, they ask AI;
  298. 9:45if they don't know something, they
  299. 9:46consult AI. And besides, ChatGPT—I've
  300. 9:48used it a few times myself—it says some
  301. 9:50really great things, doesn't it?
  302. 9:51>> Yeah.
  303. 9:52>> Like, "Your idea is wonderful."
  304. 9:53Something like that. It feels like
  305. 9:55something is missing in their hearts.
  306. 9:56>> Yeah. Like calling it Chappie, you know
  307. 9:58.
  308. 9:58>> Yeah, Chappie and such.
  309. 10:00>> That kind of vibe, right? Exactly.
  310. 10:03And for that generation, at the end of
  311. 10:05the day, judging whether what the AI
  312. 10:07says is right or wrong, or maybe it’s
  313. 10:08correct but just not interesting...
  314. 10:12>> Yeah.
  315. 10:13...
  316. 10:13>> refining your sense or things like that
  317. 10:15can’t really be done unless you engage
  318. 10:17with the world outside of AI, can it?
  319. 10:19>> Yeah. I mean, honestly, for example,
  320. 10:22book lovers would understand whether a
  321. 10:23book is interesting or not—people who go
  322. 10:26to bookstores or libraries, look at
  323. 10:27various shelves, and touch the books
  324. 10:29themselves are the ones who can...oh,
  325. 10:31this book seems interesting...
  326. 10:33>> Yeah. ...
  327. 10:33>> they can spot things like that.
  328. 10:35>> I think it’s the same with clothes,
  329. 10:36isn't it?
  330. 10:37>> Trying on different clothes yourself,
  331. 10:39going around to different shops—that’s
  332. 10:41how you hone your sense, and I think
  333. 10:43that goes for everything.
  334. 10:45And by spending your own time like that
  335. 10:47—you know, in economics, there’s the
  336. 10:49labor theory of value, and the idea that
  337. 10:52value is born from labor is a truth you
  338. 10:54see in almost any field.
  339. 10:55>> Yeah.
  340. 10:58If you spend a lot of time, do a lot of
  341. 11:00useless things, build up your experience
  342. 11:02, and in the process of interacting with
  343. 11:05others, have your sense criticized and
  344. 11:07criticize back—through that
  345. 11:09back-and-forth communication, if someone
  346. 11:11who has their own core axis of value
  347. 11:13uses AI, then that synergy could create
  348. 11:15something interesting, right?
  349. 11:19>> Well, I think that’s fine. But for young
  350. 11:21people today to do that, they really
  351. 11:23need to spend a certain amount of time
  352. 11:25not using AI or computers.
  353. 11:27>> Yeah.
  354. 11:29>> If they don't use that time to refine
  355. 11:31what’s currently called "non-cognitive
  356. 11:33skills," I feel like they’ll end up
  357. 11:35being swayed—or worse, they won't even
  358. 11:37realize they're being swayed.
  359. 11:40>> When we talk about "sense," there’s an
  360. 11:41image that it’s something you're born
  361. 11:43with, but that’s not actually the case,
  362. 11:45is it?
  363. 11:45>> Yeah. I think so. Well, of course, that
  364. 11:48aspect exists too. But I wonder about
  365. 11:50that. For example, coming up with
  366. 11:52something interesting, well, of course,
  367. 11:54a part of it is having a sense for it.
  368. 11:57You know, for instance, in university
  369. 11:59seminars, there are always people who
  370. 12:01share interesting opinions during
  371. 12:02discussions.
  372. 12:03>> Yes.
  373. 12:04>> Well, that type of person, someone who
  374. 12:06is "naturally smart," it’s not just
  375. 12:07about them reading a lot of books. I
  376. 12:09think there’s something more to it.
  377. 12:13Well, that may be, but it seems it's not
  378. 12:15really well understood yet why people
  379. 12:17become what we call "naturally smart."
  380. 12:22>> At least, that’s what I hear when I ask
  381. 12:24scholars or education experts. Well,
  382. 12:26putting that aside, yeah.
  383. 12:28>> I mean, if things were decided by sense
  384. 12:30alone, humans wouldn't be as complex as
  385. 12:32they are.
  386. 12:32>> Yeah.
  387. 12:33>> So, I think things that are learned
  388. 12:35later in life are quite significant.
  389. 12:39I really do think that plays a big role
  390. 12:41.
  391. 12:42I’ve done quite a few interviews for*
  392. 12:44Criterion*, for example, with
  393. 12:45craftspeople and the like.
  394. 12:47>> Yeah.
  395. 12:48>> I’ve interviewed them, and...
  396. 12:50>> Most of them say that as they continued
  397. 12:52their work for a long time, their talent
  398. 12:54blossomed, and they felt it was their
  399. 12:56calling, even though it wasn't like that
  400. 12:58at all in the beginning.
  401. 13:00>> Most human abilities, well, the starting
  402. 13:02points may differ, of course.
  403. 13:06But I think how things turn out later
  404. 13:08depends on the accumulation of
  405. 13:10experience and what one learns from it.
  406. 13:13This isn't just me being polite; I truly
  407. 13:15believe that.
  408. 13:16>> Yeah.
  409. 13:17>> So, bringing this back to the theory of
  410. 13:19education,
  411. 13:21>> the thing that AI relies on is
  412. 13:23experience, isn't it? Yeah. Hmm. It’s
  413. 13:25not just the experience itself, but if
  414. 13:27you ask ChatGPT about what to learn from
  415. 13:29experience,
  416. 13:30>> I feel like you’ll only get generic
  417. 13:31answers back.
  418. 13:32>> Yeah.
  419. 13:33>> That’s what I suspect.
  420. 13:35I mean, how should I put it?
  421. 13:39When humans learn from past experiences
  422. 13:42, those experiences leave a strong
  423. 13:44imprint on their memory—in philosophical
  424. 13:47terms, we might call it an "
  425. 13:48eventfulness" or "incidentality," that
  426. 13:51kind of searing memory.
  427. 13:55It could be encountering something in a
  428. 13:57truly memorable situation, or perhaps
  429. 13:59just a casual remark from a friend that
  430. 14:01sticks with you for some reason.
  431. 14:03>> Yes.
  432. 14:07>> I think experiences and memories are
  433. 14:09things that are forged as we bump into
  434. 14:11things in life and navigate through
  435. 14:12those ups and downs.
  436. 14:16>> Yes.
  437. 14:18To put it very broadly, unexpected
  438. 14:20events, failures, and surprises are what
  439. 14:22tend to remain in our memory.
  440. 14:25>> Mm.
  441. 14:26>> So, conversely, if there are no
  442. 14:28unexpected events, it's easy to forget.
  443. 14:31When I was young, I used to go
  444. 14:32backpacking alone overseas, and those
  445. 14:35experiences really stay with me.
  446. 14:37>> Mm.
  447. 14:37>> But for instance, when I go on business
  448. 14:39trips that are more like package tours,
  449. 14:41where the schedule is pretty much set,
  450. 14:43>> I end up forgetting almost every single
  451. 14:44one of them. It's true.
  452. 14:46>> Yeah. Even though it's the same overseas
  453. 14:48experience, it just doesn't sink in at
  454. 14:50all.
  455. 14:50>> Yeah.
  456. 14:51>> So, it really comes down to those types
  457. 14:53of events.
  458. 14:54Mm.
  459. 14:55>> If there's some kind of significance to
  460. 14:57it, that's what stays in your memory. I
  461. 14:58suppose humans have a brain function
  462. 15:00that tries to increase survival rates by
  463. 15:03remembering unique experiences among
  464. 15:05those events.
  465. 15:05>> Mm. Yeah.
  466. 15:07And since those things remain, I think
  467. 15:09that by ruminating on those events, the
  468. 15:11core of a person's important thoughts or
  469. 15:13their sense of sensibility is formed.
  470. 15:16>> Mm.
  471. 15:17>> You know, if you're always on a PC or
  472. 15:19using AI, it's hard to come across those
  473. 15:21kinds of incidents, or interesting
  474. 15:23things, or unexpected events, or even
  475. 15:25failures. Yes.
  476. 15:26>> That is one thing, and
  477. 15:28>> even if you do encounter an event, if
  478. 15:30you get consoled by ChatGPT, it doesn't
  479. 15:32really sink in or deepen your
  480. 15:34perspective.
  481. 15:37As someone who doesn't use AI, I feel
  482. 15:39like that might be the case.
  483. 15:43>> Well, maybe people who use AI would say
  484. 15:46, "No, we have those experiences too."
  485. 15:50But still, humanity has created
  486. 15:52something truly incredible.
  487. 15:54>> I think this is going to be quite a big
  488. 15:56deal.
  489. 15:57>> In other words, I think this touches on
  490. 15:59the things we might lose because of AI.
  491. 16:02>> Yeah.
  492. 16:05What should we call that, then?
  493. 16:08>> I wonder what it is. Something like...
  494. 16:10>> I think it’s probably things necessary
  495. 16:12for personality development or enriching
  496. 16:14one’s humanity, as the professor was
  497. 16:16saying.
  498. 16:17>> Yeah. Yeah. Yeah.
  499. 16:19>> That’s right.
  500. 16:20It’s a cliché because it’s been said for
  501. 16:23so long, but I wonder if we’re heading
  502. 16:25toward a situation where life just
  503. 16:26passes us by.
  504. 16:28>> Yeah.
  505. 16:30I get the feeling that it becomes
  506. 16:31difficult to leave the little world of
  507. 16:33comfort we’ve created for ourselves.
  508. 16:36Yeah.
  509. 16:36>> Well, we were told this kind of thing
  510. 16:38when we were kids, too. Like "you just
  511. 16:39play video games all the time."
  512. 16:41>> And the next generation was probably
  513. 16:42told they were "just on the internet all
  514. 16:43the time." Now, they’re probably being
  515. 16:44told they "just use AI all the time," so
  516. 16:46it’s been happening for a long time.
  517. 16:47>> As civilization has advanced, especially
  518. 16:49with the digital age, everyone started
  519. 16:51doing this, looking at their screens
  520. 16:53instead of the world. Maybe even further
  521. 16:55back, it was "you just read books all
  522. 16:57the time." Or just reading stories; even
  523. 17:00in the 19th century, novels were
  524. 17:02considered harmful.
  525. 17:03>> People would say, "If you keep reading
  526. 17:05those, you won't have a normal life; go
  527. 17:07interact with people, stop just reading
  528. 17:09books."
  529. 17:10>> Yeah.
  530. 17:10>> Apparently, that was a thing back in the
  531. 17:1219th century, too.
  532. 17:13>> Huh.
  533. 17:14>> So this has been around since the early
  534. 17:16modern era, but I feel like it’s
  535. 17:18gradually getting worse.
  536. 17:20Like...
  537. 17:22>> Nowadays, if you’re not careful, you can
  538. 17:25go 24 hours without interacting with
  539. 17:27anyone, barely leaving the comfortable
  540. 17:29environment you've built, and just have
  541. 17:31entertainment...
  542. 17:32>> Yeah.
  543. 17:33>> And you can get all your food at a
  544. 17:34convenience store, you can do everything
  545. 17:36. All you have to do is work, right?
  546. 17:40In that sense, there are so many things
  547. 17:42available now that make it possible to
  548. 17:44build a cozy, comfortable living space.
  549. 17:47So, I think it’s less about the person
  550. 17:49being at fault and more about the
  551. 17:51environment making things that way.
  552. 17:54Still, you know...
  553. 17:57There was this student who returned from
  554. 17:59studying in, well, not North America,
  555. 18:01but Northern Europe—Denmark.
  556. 18:03>> Yes.
  557. 18:05As you know, Denmark is a welfare state
  558. 18:07with an extremely well-developed social
  559. 18:09security system.
  560. 18:12It’s hard to find the right word, but
  561. 18:14what people used to call vagrants—or
  562. 18:16what’s the correct term now?
  563. 18:19Homeless people?
  564. 18:20>> They don't really have the right to be
  565. 18:22homeless, because society immediately
  566. 18:24swoops in to take care of them or whisk
  567. 18:26them away.
  568. 18:27>> On the flip side, though, I heard that
  569. 18:29because Denmark’s welfare is so thick,
  570. 18:31people are actually fragile when it
  571. 18:32comes to failure.
  572. 18:33>> Yeah. They’re prone to despair easily,
  573. 18:36and paradoxically, the number of
  574. 18:37homeless people increases because they
  575. 18:39just want to be left alone.
  576. 18:41>> Oh.
  577. 18:42>> There are people who say, "Don't take
  578. 18:44care of everything for me," and I can
  579. 18:46see why.
  580. 18:49Well, humans really don't feel like
  581. 18:51they're truly living unless they feel a
  582. 18:53sense of freedom in a real way, I think
  583. 18:56.
  584. 18:57>> Yeah.
  585. 18:57>> Actually, in an overly perfect
  586. 18:59environment, it's hard for people to
  587. 19:01feel that freedom.
  588. 19:02>> Yeah. When you don't feel free, your
  589. 19:04spirit begins to lose its vitality.
  590. 19:10It may be inconvenient, or even a
  591. 19:12tiresome and stressful environment, but
  592. 19:14when you encounter things there,
  593. 19:16experience them, and learn from them,
  594. 19:22people tend to feel free more easily.
  595. 19:24Yeah. And whether or not you feel that
  596. 19:26freedom makes a big difference in how
  597. 19:27fulfilling your life is.
  598. 19:28>> Yes.
  599. 19:29>> So, in that sense, humanity won't just
  600. 19:32be wiped out by AI.
  601. 19:34>> Somewhere along the line, seeking that
  602. 19:36freedom I just mentioned,
  603. 19:38>> people will likely head back out into
  604. 19:40the world of real-life experiences,
  605. 19:42don't you think? I’m half-optimistically
  606. 19:44hoping that even the children doing this
  607. 19:46now will eventually turn out that way.
  608. 19:48>> Yeah. Once you step out into society,
  609. 19:49you’ll inevitably face situations where
  610. 19:51you have to deal with people, like it or
  611. 19:52not.
  612. 19:52>> Yeah. Yeah. Yeah. And frankly, looking
  613. 19:54at the results, there are actually a lot
  614. 19:56more interesting things that happen that
  615. 19:58way.
  616. 19:58>> Yeah. Because unexpected things happen
  617. 20:00when you deal with other people.
  618. 20:01>> Yeah. And well, even things that make
  619. 20:04you fed up or troubles you encounter,
  620. 20:06when some time passes, you realize they
  621. 20:08actually became something important in
  622. 20:10your life.
  623. 20:11>> Yeah.
  624. 20:12>> And if you're able to think that way, it
  625. 20:14means you've been managing your life
  626. 20:16quite well since then.
  627. 20:18>> Yeah.
  628. 20:20I think looking back and confirming how
  629. 20:22you've managed your life in that way is
  630. 20:24probably a part of life's joy, isn't it
  631. 20:26?
  632. 20:28>> I think so, too.
  633. 20:30>> So, finally, this issue of The Criterion
  634. 20:33features a special on AI—could you give
  635. 20:35us a word on that, Professor?
  636. 20:38>> Yeah.
  637. 20:40In this AI feature, we didn't just have
  638. 20:42people from the humanities or philosophy
  639. 20:45, but also experts from other fields
  640. 20:47contribute, making it a very substantial
  641. 20:49read.
  642. 20:51>> Yeah.
  643. 20:52>> I wonder if Professor Kawa translated
  644. 20:54this? This...excuse me, I can't quite
  645. 20:57recall the name. Ah, you mean Marshall
  646. 21:00Ford.
  647. 21:01I thought his argument was really
  648. 21:02interesting.
  649. 21:05It contains many different points of
  650. 21:06discussion, so I'm sure it will
  651. 21:08stimulate your thinking.
  652. 21:11>> It’s full of things not written in
  653. 21:13ChatGPT, so please do pick up a copy.
  654. 21:16>> Yes.
  655. 21:18With that, thank you again for your time
  656. 21:20today.
  657. 21:21>> The Monthly Hyogensha official LINE
  658. 21:23account delivers various information,
  659. 21:26from new book discount notices and video
  660. 21:28announcements to exclusive LINE content
  661. 21:30. Add us as a friend via the QR code or
  662. 21:33from the description box.

About this transcript

This page contains the full transcript of 【AI時代の教育論】卒業論文がなくなる?レポート試験は意味がない?AIと大学の最前線(解説:京都大学准教授 柴山桂太) by 『表現者クライテリオン』チャンネル, generated from the public captions YouTube serves with the video. The transcript has 3,692 words across 662 segments, with the original timestamps preserved so you can click any line to jump to that moment in the embedded player.

What you can do with it

Use the transcript to take notes, quote the speaker, build a study guide, generate a summary with ChatGPT or Claude via the YouTube Summary tool, or export it as a timed subtitle file with YouTube to SRT. You can also re-open it in the transcriber to choose a caption track available on the video.

Free YouTube transcript tool

YouTube2Text is a free YouTube transcript generator — no signup, no daily limit. Paste any YouTube link and get the full transcript instantly, with timestamps, click-to-jump, caption language selection, AI prompts for ChatGPT, Claude, and Gemini, and exports to TXT, SRT, VTT, or Markdown.