YouTube2Text

AI in Schools: Cheater or Tutor? | Paul Matthews | TEDxHobart — Transcript

by TEDx Talks · 3,031 words · 458 segments · language en · Watch on YouTube

Full transcript

  1. 0:00[Music]
  2. 0:04[Applause]
  3. 0:04[Music]
  4. 0:10I want to introduce you to Jess a year n
  5. 0:13student from right here in Hobart now
  6. 0:16Jess is a lovely girl but unfortunately
  7. 0:18she really struggles at school you see
  8. 0:21Jess struggles with the basics things
  9. 0:23like reading and writing math she gets a
  10. 0:27report card home every year and on it it
  11. 0:29says she's a approaching the standard
  12. 0:30but in reality Jess is falling further
  13. 0:34and further behind with each passing
  14. 0:36year she really struggles to do her
  15. 0:38classwork she hates putting up her hand
  16. 0:40all the time and asking for help and
  17. 0:42well because her teachers got 28 other
  18. 0:44students in the
  19. 0:45room Jess is slipping between the
  20. 0:48cracks now friends the thing about
  21. 0:50Jess's situation is that it's entirely
  22. 0:54preventable Jesse is struggling right
  23. 0:56now but we could actually get her to be
  24. 0:58a high achiever before the years
  25. 1:00out the research shows us that all we
  26. 1:03really need for
  27. 1:04Jess is one-on-one full-time support
  28. 1:08that's all we need Jess needs a tutor
  29. 1:11right so if we were able to put our
  30. 1:13wallets together our heads together
  31. 1:15maybe we launch a crowdfunding campaign
  32. 1:18we can get the $80,000 necessary to get
  33. 1:20Jess a full-time tutor and that's going
  34. 1:23to change the shape of her education
  35. 1:25even her life forever
  36. 1:30the only catch with that
  37. 1:31plan is that in Tasmania Jess's
  38. 1:34struggles struggling with the basics
  39. 1:37well they're actually really common so
  40. 1:39if we zoom out from the person and we
  41. 1:41look at the whole problem actually we
  42. 1:42don't need to raise $80,000 we need to
  43. 1:45raise $80,000 lots of
  44. 1:48$80,000 and I actually I don't fancy our
  45. 1:51chances of raising 6.4 billion dollars
  46. 1:54in our crowdfunding
  47. 1:56campaign so while a one-on-one tter full
  48. 1:59time support for each of our Tasmanian
  49. 2:01students well it would reverse the
  50. 2:03fortunes of our struggling education
  51. 2:06system it's unfortunately just not
  52. 2:08possible this can't be done but what I'm
  53. 2:11here to tell you today friends is that
  54. 2:12in a world infused with artificial
  55. 2:14intelligence we can get many of those
  56. 2:16same benefits for our students right now
  57. 2:20and today I'm going to show you how so
  58. 2:22my name is Paul Matthews and I'm
  59. 2:24passionate about helping students
  60. 2:26teachers and school communities not just
  61. 2:28survive but thrive in a world infused
  62. 2:31with artificial
  63. 2:32intelligence so education is in My DNA
  64. 2:35I'm a third generation
  65. 2:38teacher and now I help students teachers
  66. 2:41and school communities thrive in an AI
  67. 2:43world and I do that as a
  68. 2:45speaker I do that as a consultant and I
  69. 2:48do that as CEO of my teacher a a tech
  70. 2:51company aimed at bringing userfriendly
  71. 2:53Ai and putting it in the hands of every
  72. 2:56teacher so as you can see I'm really
  73. 2:58sold on the possib abilities of AI and
  74. 3:02education perhaps as you come here
  75. 3:04though you're not
  76. 3:06sold perhaps you're a bit of an AI
  77. 3:09skeptic maybe you're even a tech skeptic
  78. 3:11in general you know you not always
  79. 3:13looking up the reviews on the latest
  80. 3:15phones you're not buying the newest
  81. 3:17watches maybe although you don't
  82. 3:19publicize it too widely you think you'll
  83. 3:20get through your life just fine without
  84. 3:22a smart
  85. 3:28fridge if that's you I want to tell you
  86. 3:31that even you friend even you should be
  87. 3:33excited about artificial intelligence
  88. 3:35and education and here's a key idea I
  89. 3:37want you to grab this you should be
  90. 3:39excited about Ai and education because
  91. 3:41the impact is much more exciting than
  92. 3:42the tech the impact is more exciting
  93. 3:45than the tech the technology things like
  94. 3:46large language models and machine
  95. 3:49learning neural networks if you don't go
  96. 3:51in for that sort of thing that's fine
  97. 3:52don't worry about that because I know
  98. 3:54what does get you
  99. 3:55excited seeing Jess and the tens of
  100. 3:59thousands of Tasmanian students like her
  101. 4:01succeed at school see friends if you get
  102. 4:04excited at seeing Tasmanian students get
  103. 4:07better grades have more fun at school
  104. 4:10and Foster a love of lifelong learning
  105. 4:11then let me tell you you should be
  106. 4:13excited about artificial intelligence
  107. 4:15and
  108. 4:16education so let me show you one of the
  109. 4:18key pieces of research that undergirds
  110. 4:21my optimism for AI and education so the
  111. 4:24idea that we could take a struggling
  112. 4:26student and help turn them into a high
  113. 4:29achiever
  114. 4:30that's not just a nice sounding idea
  115. 4:32it's actually based on one of the
  116. 4:33seminal pieces of educational research
  117. 4:35to come out of the 20th century now in
  118. 4:381984 Professor Benjamin Bloom he
  119. 4:40published a paper called the two Sigma
  120. 4:43problem now in this paper he set out to
  121. 4:46find how effective different kinds of
  122. 4:49teaching was so he compared a
  123. 4:52conventional class with people who were
  124. 4:54cheed oneon-one now a conventional class
  125. 4:58it's what I had it's probably what you
  126. 5:00had it's where there's one teacher and
  127. 5:0220 or 30 students right and what did
  128. 5:06Benjamin Bloom find when he compared his
  129. 5:08conventional class to one-on-one tuted
  130. 5:10students well this is what he found and
  131. 5:13you can see it on the graph right there
  132. 5:14that normally distributed bell curve
  133. 5:16that was a conventional class it's
  134. 5:19exactly what you'd expect to
  135. 5:21see but you can see the curve that's
  136. 5:24pushed up to the higher end of the
  137. 5:26scale that's the performance of the
  138. 5:28tuted
  139. 5:30students as you can see the tuted
  140. 5:32students it made an eyew watering
  141. 5:34difference for them let me give you two
  142. 5:37pieces of data to put legs on
  143. 5:40that first of all if you're at the
  144. 5:43average tuted student just dead average
  145. 5:45middle of the cohort you fell above
  146. 5:4998% of the conventional class that's an
  147. 5:52eyeing
  148. 5:53difference second of
  149. 5:55all students who were chewed
  150. 5:58one-on-one on a standardized test they
  151. 6:02perform two standard deviations higher
  152. 6:04than the conventional
  153. 6:06class so as you can see here oneon-one
  154. 6:08tutoring it makes an eye watering
  155. 6:11lifechanging two standard deviation
  156. 6:14difference in a student's
  157. 6:16life now for those of you who've got
  158. 6:18your Math's hat on you'll know that a
  159. 6:20standard deviation in the math
  160. 6:21literature is called a sigma it's called
  161. 6:24a sigma and that's why this paper was
  162. 6:26entitled the two Sigma problem right two
  163. 6:29standard deviation increase in learning
  164. 6:32two
  165. 6:33Sigma but the real question here is why
  166. 6:35was it called the two Sigma
  167. 6:38problem what's the big problem with
  168. 6:40getting people so much
  169. 6:43learning well friends the problem is we
  170. 6:47just can't get those eye watering
  171. 6:50lifechanging two standard deviation
  172. 6:53differences for all our students we just
  173. 6:54don't have the
  174. 6:55resources so unfortunately for Jess and
  175. 6:58the tens and thousands of Tasmanian
  176. 7:00students like her the hypothetical
  177. 7:02benefits of that kind of approach well
  178. 7:03they remain just that
  179. 7:06hypothetical that's a two Sigma
  180. 7:10problem but what I'm here to tell you
  181. 7:12today friends is that in a world infused
  182. 7:15with artificial intelligence the two
  183. 7:18Sigma problem has become a two Sigma
  184. 7:21possibility we can use artificial
  185. 7:23intelligence right now to put students
  186. 7:25in a more tailored personalized feedback
  187. 7:28Rich environment than ever
  188. 7:31before of course when AI first came out
  189. 7:34or at least when it hit the mainstream
  190. 7:35in November
  191. 7:372022 the dominant narrative was that
  192. 7:39students would use it to cheat I wonder
  193. 7:41if you heard that one I certainly I was
  194. 7:43in a school at the time Educators were
  195. 7:45running around lighting their hair on
  196. 7:46fire this is the end of Education this
  197. 7:48is the death of learning they
  198. 7:50said and because of that knee-jerk fear
  199. 7:54response well people just didn't
  200. 7:57see that actually AI might be exactly
  201. 8:00what we need for our tied teachers and
  202. 8:02our struggling students see people just
  203. 8:05didn't pick up on the fact that maybe we
  204. 8:07could use this new technology to put our
  205. 8:09students in a more tailored personalized
  206. 8:11feedback-rich environment than ever
  207. 8:13before possible of course AI could be
  208. 8:16bad of course it could be used poorly
  209. 8:19but it could also be fantastic it could
  210. 8:21also be just what we need and that's why
  211. 8:23I'm working as hard as I can to make
  212. 8:25sure AI is a benefit and a blessing to
  213. 8:27our students and our teachers
  214. 8:31so today friends I want to give you
  215. 8:32three simple strategies that we can put
  216. 8:34into place right now that will have a
  217. 8:37huge impact on Jess and the tens of
  218. 8:40thousands of Tasmanian students like
  219. 8:43her now the interesting thing you'll
  220. 8:45note about these strategies is they're
  221. 8:46not brand new ideas they're not big
  222. 8:49Innovations they're actually things
  223. 8:50teachers have been doing for a really
  224. 8:52long time they're things that I did in
  225. 8:53my classroom before
  226. 8:56AI it's a little bit of a window into my
  227. 8:58vision for artificial intelligence in
  228. 9:01education it's not about brand new ideas
  229. 9:03it's about doing the basics better with
  230. 9:06the AI doing the things that we know
  231. 9:08work more often for more of our
  232. 9:11students and that's exactly what you're
  233. 9:13going to see in these three simple
  234. 9:15strategies now you know when you engage
  235. 9:18with AI and you ask it to do something
  236. 9:21for you that's called a prompt now I've
  237. 9:24written up prompts for these three
  238. 9:26simple strategies that will have a huge
  239. 9:27impact they didn't fit on my slide
  240. 9:30so you'll note on my last slide I've
  241. 9:31actually got a QR code there what I want
  242. 9:34you to do when that comes up if you're
  243. 9:35able to scan that that'll be on the last
  244. 9:37slide you'll be able to download my
  245. 9:39ebook that contains a simple framework
  246. 9:41for how you can use AI as well as the
  247. 9:44prompts for these three simple
  248. 9:47strategies so let's dive in now and have
  249. 9:50a look at some simple strategies that
  250. 9:51will have a huge impact for Jess and the
  251. 9:53tens of thousands of Tasmania students
  252. 9:55like her the first one is text
  253. 9:58differentiation now that's it's a big
  254. 10:00phrase but what it means is the teacher
  255. 10:03adjusting the readability of a text
  256. 10:06maybe reducing the length or the
  257. 10:08complexity for people in the class who
  258. 10:10don't read so well that's text
  259. 10:13differentiation because you see friends
  260. 10:15in Jess's year n class there are people
  261. 10:18reading at a grade n level and that's
  262. 10:20good there are some people who are
  263. 10:21reading at a grade seven
  264. 10:23level some at a grade five
  265. 10:26level some at a grade three level
  266. 10:30so if the teacher is handing everyone
  267. 10:32the same worksheet and asking them to
  268. 10:33read it that's what I like to call the
  269. 10:34spray and pray approach you're just
  270. 10:36spraying out the same resource to
  271. 10:37everyone and you're just praying they
  272. 10:39can read
  273. 10:41it in a world infused with AI we can do
  274. 10:44a lot better for our Learners right now
  275. 10:46it really is as simple as a teacher
  276. 10:48taking the text that they're going to
  277. 10:50read in class you plug that into an AI
  278. 10:53tool like chat GPT or my teacher ID and
  279. 10:56you say can you please reduce the length
  280. 10:58of comple and complexity of
  281. 11:00this so they bump it down from a grade
  282. 11:02N9 level to a grade seven level say can
  283. 11:04you please reduce the length and
  284. 11:06complexity further bumps it down to a
  285. 11:08grade five level please reduce this to
  286. 11:10six simple dot points you've got your
  287. 11:13grade three version right there it has
  288. 11:16never been easier to make sure every
  289. 11:19Learner in our classroom can read what
  290. 11:22we're asking them to read that's a
  291. 11:25simple strategy but friends it will have
  292. 11:26a huge impact it's not a brand new idea
  293. 11:29just doing the basics better with
  294. 11:32AI the second
  295. 11:34strategy that we can put in place it
  296. 11:36will have a huge impact is called
  297. 11:38retrieval practice let me unpack what
  298. 11:40that means we used to think that
  299. 11:43teaching a student was all about what
  300. 11:44you could put in their head it's like
  301. 11:46stuffing a sleeping bag into a sack just
  302. 11:48got to put that knowledge in
  303. 11:51there what we're realizing now is it's
  304. 11:53not so much about what you can put in
  305. 11:55although that's really important it's
  306. 11:56also really important to draw it out as
  307. 11:58well you got to draw that knowledge out
  308. 12:00of the student's head and that process
  309. 12:02of drawing it out that's called
  310. 12:03retrieval practice you see much like a
  311. 12:06student needs to practice an instrument
  312. 12:08or a sport they actually need to
  313. 12:09practice their knowledge as well now
  314. 12:13teachers have always used a really
  315. 12:14simple quiz for this we use a quiz so
  316. 12:16you read something and you get quizzed
  317. 12:17on it great way to practice your
  318. 12:19knowledge we just don't do it that often
  319. 12:21because well those quizzes take a long
  320. 12:23time to
  321. 12:25create what we can do now using AI we
  322. 12:28can allow students to practice their
  323. 12:30knowledge without the huge time penalty
  324. 12:32on the teachers it really is as simple
  325. 12:34as taking the text that you're reading
  326. 12:36or the transcript of the YouTube video
  327. 12:38that you're looking
  328. 12:39at you plug it into an AI tool like chat
  329. 12:42GPT or my teacher Aid and you say please
  330. 12:45create me four multiple choice questions
  331. 12:47four short answer questions four fill-in
  332. 12:48theblank questions and you've got it
  333. 12:51right there so we can get our students
  334. 12:54practicing their knowledge but it's not
  335. 12:55costing our teachers their evenings
  336. 12:57weekends and holidays
  337. 13:00you see it's a simple strategy but it
  338. 13:01will have a huge impact it's not a brand
  339. 13:05new idea it's just doing the basics
  340. 13:06better with
  341. 13:09AI the third simple strategy that will
  342. 13:11have a huge impact is training students
  343. 13:14to get feedback on their work from
  344. 13:16artificial
  345. 13:17intelligence feedback is crucial for
  346. 13:20anyone who's trying to learn anything
  347. 13:22right and it's it's a really clear
  348. 13:23advantage of a one-on-one tutor over a
  349. 13:26conventional class isn't it one-on-one
  350. 13:28tutoring it just allows way more space
  351. 13:30and time for critique or course
  352. 13:32correction or
  353. 13:34encouragement as a teacher in a
  354. 13:36conventional classroom for many years I
  355. 13:38know there are going to be some days
  356. 13:39maybe even some weeks or months where
  357. 13:42students don't get that tailored
  358. 13:44personalized feedback from
  359. 13:47us we can train students so that instead
  360. 13:50of using AI to do their work for
  361. 13:53them they can use it to help them with
  362. 13:55their work so instead of saying write my
  363. 13:57essay please
  364. 13:59they can say well here is my essay can
  365. 14:01you please grade it please give me some
  366. 14:03feedback about logical coherence and
  367. 14:05structure and
  368. 14:07vocab now if you're a teacher out there
  369. 14:09it's worth saying this this is not about
  370. 14:11replacing the teachers absolutely not
  371. 14:14teachers cannot be
  372. 14:17replaced I always say to teachers
  373. 14:19whenever I'm speaking to them AI is not
  374. 14:21here to take your job it's here to help
  375. 14:22you do a better job fact of the matter
  376. 14:24is our students need feedback we're
  377. 14:27limited to being at one place one time
  378. 14:30and we can put students in that tailored
  379. 14:32personalized feedback Rich environment
  380. 14:35using
  381. 14:36AI so those are three simple strategies
  382. 14:39that we could do tomorrow and it would
  383. 14:42have a huge impact on Jess and the tens
  384. 14:45of thousands of Tasmanian students like
  385. 14:48her but if AI is going to be a benefit
  386. 14:52and a blessing to us we need to approach
  387. 14:53it carefully it's not the sort of thing
  388. 14:55we're going to get right by accident
  389. 14:56here are two things we have to do first
  390. 14:58of all teachers need training and they
  391. 15:00need time off they need high quality
  392. 15:03professional development and they need
  393. 15:04time to be able to learn how to use
  394. 15:06these
  395. 15:07tools you
  396. 15:09see AI is the biggest disruption
  397. 15:11education has ever seen it's bigger than
  398. 15:14the printing press it's bigger than Co
  399. 15:18we can't expect Educators to try and
  400. 15:20wrap their head around this off the side
  401. 15:22of their desk right AI can't just be
  402. 15:24another thing on the long list of things
  403. 15:26teachers are trying to get done in their
  404. 15:27spare time
  405. 15:29we need to give them training and time
  406. 15:32off to learn how to use these tools
  407. 15:34second of all we have to train our
  408. 15:36students to use AI we absolutely have to
  409. 15:39the reason for that is when it comes to
  410. 15:41AI use in students it's not a matter of
  411. 15:44if it's a matter of How It's not if
  412. 15:47they'll use it of course they'll use it
  413. 15:48they're already using it I see it every
  414. 15:50single day AI is everywhere we can't
  415. 15:53expect to ban it that horse is bolted or
  416. 15:56as I say to my students that toothpaste
  417. 15:58is out of the ch
  418. 15:59it's out there it can't be put back
  419. 16:02in it's not a question of if they'll use
  420. 16:04it it's a question of how will they use
  421. 16:06it in a way that hurts their education
  422. 16:08or helps their education it's that same
  423. 16:10idea that's the title of my talk AI
  424. 16:13cheater or cheor will it be a benefit or
  425. 16:17will it be harmful instead of doing the
  426. 16:19work for them we can train students to
  427. 16:22use AI to get feedback on their work and
  428. 16:25help them in their learning so those are
  429. 16:26two things we have to do if we're get
  430. 16:28going to get it right teachers need time
  431. 16:30off and training we have to show our
  432. 16:32students how to use it
  433. 16:34well so friends that's my vision for AI
  434. 16:38in education it's not about brand new
  435. 16:41ideas it's about doing the basics better
  436. 16:45with AI it's about simple strategies
  437. 16:47that will have a huge impact on
  438. 16:50Jess and the tens of thousands of
  439. 16:53Tasmanian students like
  440. 16:56her you know that today what talking
  441. 16:59about Ripple
  442. 17:00effects let me tell you something that
  443. 17:02you probably already
  444. 17:04know education is the meta ripple effect
  445. 17:08it is a ripple effect that sits above
  446. 17:10every single other ripple effect because
  447. 17:12of course it starts in the classroom but
  448. 17:14it worked its way out into every single
  449. 17:16facet of society and let me tell you
  450. 17:19this if we can use AI to do the basics
  451. 17:23of Education better do the things that
  452. 17:25we know work more often for more
  453. 17:27students if we employ these simple
  454. 17:30strategies that will have a huge impact
  455. 17:32we are going to see one of the most
  456. 17:33powerful and exciting Ripple effects
  457. 17:36throughout our state our nation and the
  458. 17:39world thank you

About this transcript

This page contains the full transcript of AI in Schools: Cheater or Tutor? | Paul Matthews | TEDxHobart by TEDx Talks, generated from the public captions YouTube serves with the video. The transcript has 3,031 words across 458 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 translate the transcript into 100+ languages.

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, translation to 100+ languages, AI prompts for ChatGPT, Claude, and Gemini, and exports to TXT, SRT, VTT, or Markdown.