WEBVTT 00:00:00.000 --> 00:00:03.339 Imagine a tool so powerful it boosts your productivity, 00:00:03.560 --> 00:00:06.820 right? But what if, kind of secretly, it dimmed 00:00:06.820 --> 00:00:09.279 your own thinking? We're diving deep into some, 00:00:09.400 --> 00:00:12.300 well, unsettling research today. It suggests 00:00:12.300 --> 00:00:14.560 our AI assistants might just be making us, you 00:00:14.560 --> 00:00:16.699 know, a little less sharp. Welcome to the Deep 00:00:16.699 --> 00:00:19.280 Dives. Today we're unpacking a really fascinating 00:00:19.280 --> 00:00:21.359 and honestly somewhat provocative study from 00:00:21.359 --> 00:00:23.780 MIT. It's got this title, Your Brain on Chat 00:00:23.780 --> 00:00:27.480 GPT. Accumulation of Cognitive Debt. when using 00:00:27.480 --> 00:00:30.640 an AI assistant for essay writing tasks. That 00:00:30.640 --> 00:00:32.939 title, it grabbed us immediately. Oh, absolutely. 00:00:33.200 --> 00:00:35.240 Cognitive debt. That phrase, it just sticks, 00:00:35.359 --> 00:00:37.200 doesn't it? It instantly brings up this idea 00:00:37.200 --> 00:00:39.240 that maybe we're borrowing, maybe without even 00:00:39.240 --> 00:00:41.700 realizing it, from our future mental abilities. 00:00:42.189 --> 00:00:44.289 So in this deep dive, we'll explore exactly what 00:00:44.289 --> 00:00:47.469 that means, how a pretty ingenious experiment 00:00:47.469 --> 00:00:49.490 uncovered these effects, which are surprising, 00:00:49.689 --> 00:00:53.170 and crucially, how we can use AI intelligently 00:00:53.170 --> 00:00:55.189 without losing our own edge. Yeah, our mission 00:00:55.189 --> 00:00:56.969 today is to understand not just the specific 00:00:56.969 --> 00:00:59.030 findings, but really grapple with the bigger 00:00:59.030 --> 00:01:01.270 picture. What does this mean for how we learn, 00:01:01.369 --> 00:01:04.250 how we think, how we deal with information now 00:01:04.250 --> 00:01:07.689 that AI is everywhere? Okay, so let's unpack 00:01:07.689 --> 00:01:10.609 this. The author of our source material, Max 00:01:10.609 --> 00:01:12.790 Ann, he describes having a bit of a personal 00:01:12.790 --> 00:01:16.049 crisis after reading this study. Like a lot of 00:01:16.049 --> 00:01:18.989 us, I think, he saw these AI writing tools as... 00:01:19.549 --> 00:01:22.409 They're all superpowers, game changers. Totally. 00:01:22.510 --> 00:01:24.629 I mean, think about it. Staring at that blank 00:01:24.629 --> 00:01:26.609 page, the blinking cursor, it's intimidating. 00:01:26.810 --> 00:01:28.650 And then you find this AI, you type, you know, 00:01:28.650 --> 00:01:30.769 write an intro about financial reporting and 00:01:30.769 --> 00:01:33.590 boom, it's just there. Perfect paragraph. The 00:01:33.590 --> 00:01:35.769 relief is instant. It feels kind of like magic. 00:01:35.890 --> 00:01:37.769 Right. And that exact feeling, that ease, that 00:01:37.769 --> 00:01:39.870 relief is what the MIT researchers were digging 00:01:39.870 --> 00:01:42.409 into. They found pretty quickly that using AI 00:01:42.409 --> 00:01:44.989 significantly lowered what they called cognitive 00:01:44.989 --> 00:01:47.170 load, which just means your brain didn't have 00:01:47.170 --> 00:01:49.459 to work nearly as hard to get the job done. Less 00:01:49.459 --> 00:01:51.799 effort. Yeah, it's the classic convenience trap, 00:01:52.000 --> 00:01:54.799 isn't it? Like constantly using GPS, even in 00:01:54.799 --> 00:01:57.400 a city you plan to live in for ages. Sure, it 00:01:57.400 --> 00:02:00.040 gets you there fast, efficiently. But you learn 00:02:00.040 --> 00:02:02.420 basically nothing about the actual routes, the 00:02:02.420 --> 00:02:05.840 layout. You don't build that mental map. The 00:02:05.840 --> 00:02:09.460 study strongly suggests AI can be like that GPS. 00:02:09.860 --> 00:02:12.590 But for our thoughts. for writing. We get to 00:02:12.590 --> 00:02:15.250 a finished essay, wow, super quickly, yes. But 00:02:15.250 --> 00:02:17.650 our brain often just skips the harder parts, 00:02:17.669 --> 00:02:19.650 the effort needed to really make those connections 00:02:19.650 --> 00:02:22.270 for understanding, for memory. It's a shortcut. 00:02:22.509 --> 00:02:25.599 And here's the kicker. While the AI users, yeah, 00:02:25.680 --> 00:02:27.520 they produce more words, faster, more output, 00:02:27.639 --> 00:02:29.159 the researchers found their reasoning quality 00:02:29.159 --> 00:02:32.659 was actually lower. It's like quantity over genuine 00:02:32.659 --> 00:02:34.659 deep thought. You know, building loads of identical 00:02:34.659 --> 00:02:37.360 houses fast versus carefully crafting one solid, 00:02:37.439 --> 00:02:40.020 unique one. Efficiency versus robustness. Yeah. 00:02:40.060 --> 00:02:41.819 So that leads to the core question, doesn't it? 00:02:41.879 --> 00:02:44.580 Does making tasks easier, lowering that cognitive 00:02:44.580 --> 00:02:46.819 load, does that inherently get in the way of 00:02:46.819 --> 00:02:48.900 deeper understanding, of really learning it? 00:02:49.060 --> 00:02:51.919 Yeah. Often, less effort does mean less learning. 00:02:52.810 --> 00:02:54.810 Think of your brain like a muscle, right? Lift 00:02:54.810 --> 00:02:57.509 easy weights or use a machine doing most of the 00:02:57.509 --> 00:03:00.979 work. You don't build real strength. Growth comes 00:03:00.979 --> 00:03:03.159 from the struggle, from engaging those muscles. 00:03:03.539 --> 00:03:06.180 It's the same for our minds. Less effort often 00:03:06.180 --> 00:03:09.360 means less deep processing, less actual learning 00:03:09.360 --> 00:03:12.919 that sticks. The brain has to actively build 00:03:12.919 --> 00:03:15.719 knowledge, not just receive it. Okay, so this 00:03:15.719 --> 00:03:17.719 is where the experiment itself gets really, really 00:03:17.719 --> 00:03:20.900 interesting. The researchers set up this three 00:03:20.900 --> 00:03:23.740 -way kind of mental showdown. To see these effects 00:03:23.740 --> 00:03:25.860 directly. That's right. They had three teams 00:03:25.860 --> 00:03:28.199 all writing essays on the same stuff. One was 00:03:28.199 --> 00:03:30.319 brain only, just them and their thoughts. Another 00:03:30.319 --> 00:03:32.099 was a search engine. They could look things up 00:03:32.099 --> 00:03:34.599 online. And the third group, they had a powerful 00:03:34.599 --> 00:03:36.919 AI assistant, Chet GPT in this case, helping 00:03:36.919 --> 00:03:39.120 them write. And the twist, the really clever 00:03:39.120 --> 00:03:41.780 part, was when they swapped methods later. They 00:03:41.780 --> 00:03:44.039 wanted to see, OK, are these AI effects just 00:03:44.039 --> 00:03:46.479 temporary or do they stick around? Like, can 00:03:46.479 --> 00:03:48.620 your brain bounce back after relying on that 00:03:48.620 --> 00:03:50.900 help? A test of cognitive resilience, really. 00:03:51.199 --> 00:03:54.439 And the results were, well, pretty stark, especially 00:03:54.439 --> 00:03:56.819 about memory. After they all finished their essays, 00:03:57.020 --> 00:04:00.219 they got a super simple test. Quote, one sentence, 00:04:00.419 --> 00:04:02.840 just one sentence from your own essay. Simple 00:04:02.840 --> 00:04:05.560 recall. The brain -only group and the search 00:04:05.560 --> 00:04:09.439 engine group. Almost perfect. Near flawless recall. 00:04:09.639 --> 00:04:11.800 They remembered what they wrote. Their brains 00:04:11.800 --> 00:04:14.599 were clearly engaged, you know, processing and 00:04:14.599 --> 00:04:17.399 coding. OK, but get ready for this number. A 00:04:17.399 --> 00:04:20.899 staggering 83 percent, 83 of the AI assistant 00:04:20.899 --> 00:04:23.680 group could not quote or even accurately paraphrase 00:04:23.680 --> 00:04:25.980 a single sentence from the essay they supposedly 00:04:25.980 --> 00:04:29.220 wrote. Not one accurate quote from anyone in 00:04:29.220 --> 00:04:30.980 that group. It's like the ideas just flowed through 00:04:30.980 --> 00:04:33.579 them, but didn't actually land. Wow. Let that 00:04:33.579 --> 00:04:37.120 sink in. 83 percent. They created it technically, 00:04:37.360 --> 00:04:40.660 but remembered almost none of it. It's like watching 00:04:40.660 --> 00:04:42.720 someone else work out and expecting to get stronger 00:04:42.720 --> 00:04:45.139 yourself. Your muscles know better. And apparently, 00:04:45.199 --> 00:04:47.939 so does your brain. It shows a real lack of true 00:04:47.939 --> 00:04:50.459 mental encoding. Your brain wasn't actively working 00:04:50.459 --> 00:04:52.319 that info, forming the connections needed to 00:04:52.319 --> 00:04:54.980 recall it, even if it was your work. That memory 00:04:54.980 --> 00:04:59.379 lapse is just wow. So beyond just memory, what 00:04:59.379 --> 00:05:01.319 does this really tell us about how deeply we 00:05:01.319 --> 00:05:04.259 engage with AI -generated stuff? I think it suggests 00:05:04.259 --> 00:05:06.259 a pretty fundamental shift, maybe. We become 00:05:06.259 --> 00:05:09.209 more like curators, maybe editors. but not necessarily 00:05:09.209 --> 00:05:11.569 creators in the old sense, and definitely not 00:05:11.569 --> 00:05:13.389 deep learners if the information isn't truly 00:05:13.389 --> 00:05:15.889 getting inside. And it wasn't just memory. The 00:05:15.889 --> 00:05:17.410 study also looked at this thing they called a 00:05:17.410 --> 00:05:19.970 crisis of ownership, the brain -only and search 00:05:19.970 --> 00:05:23.350 engine folks. They felt 100 % ownership, pride, 00:05:23.610 --> 00:05:26.389 connection to their work. Yeah, but the AI group, 00:05:26.529 --> 00:05:28.610 totally different story. Many felt it was like 00:05:28.610 --> 00:05:31.990 50 -50 collaboration at best. Some even felt, 00:05:32.069 --> 00:05:34.980 remarkably, No ownership at all. That feeling 00:05:34.980 --> 00:05:37.480 of this is mine, that responsibility, it just 00:05:37.480 --> 00:05:40.060 wasn't there as strongly. Faded. The researchers 00:05:40.060 --> 00:05:43.259 also found this intelligence divide in how people 00:05:43.259 --> 00:05:45.779 use the AI. Higher competence learners, they 00:05:45.779 --> 00:05:47.839 tended to use it strategically, like a sparring 00:05:47.839 --> 00:05:50.160 partner, you know, or a sharp editor. They used 00:05:50.160 --> 00:05:51.959 it to enhance their own skills, push their thinking. 00:05:52.139 --> 00:05:53.839 Right. Whereas the lower competence learners 00:05:53.839 --> 00:05:56.279 often used it very differently. They relied on 00:05:56.279 --> 00:05:59.579 it more to just. do the work for them, not really 00:05:59.579 --> 00:06:02.399 to help them learn how to do it. It's that classic 00:06:02.399 --> 00:06:05.939 tool versus crutch thing, like an engineer using 00:06:05.939 --> 00:06:08.439 a calculator for complex stuff versus a student 00:06:08.439 --> 00:06:11.500 using it to avoid basic math. One helps, one 00:06:11.500 --> 00:06:13.620 hinders learning. And they actually looked inside 00:06:13.620 --> 00:06:17.660 the brain using FNIRS. which is fascinating you 00:06:17.660 --> 00:06:19.420 measure brain activity by tracking blood flow 00:06:19.420 --> 00:06:21.279 shows what's happening under the hood exactly 00:06:21.279 --> 00:06:23.920 and they saw this really clear difference brain 00:06:23.920 --> 00:06:26.959 only and search engine users they showed bottom 00:06:26.959 --> 00:06:29.600 up processing that's like discovery mode your 00:06:29.600 --> 00:06:31.899 brain building knowledge piece by piece like 00:06:31.899 --> 00:06:34.540 putting together a puzzle from scratch it takes 00:06:34.540 --> 00:06:36.860 effort sure but that's how you really understand 00:06:36.860 --> 00:06:39.470 but the ai assistant users Totally different 00:06:39.470 --> 00:06:41.949 pattern. They showed top -down processing. It's 00:06:41.949 --> 00:06:43.670 like they were given the big picture, the finished 00:06:43.670 --> 00:06:45.649 puzzle outline. And we're mostly just filling 00:06:45.649 --> 00:06:48.670 in details, tracing lines almost, not constructing, 00:06:48.850 --> 00:06:52.629 just validating. And that meant, overall, significantly 00:06:52.629 --> 00:06:55.329 less brain activity, less connectivity in that 00:06:55.329 --> 00:06:58.029 group, less mental work, less mental growth. 00:06:58.250 --> 00:06:59.990 So wait, it's not just about what you produce. 00:07:00.029 --> 00:07:02.730 AI literally changes how your brain functions 00:07:02.730 --> 00:07:05.019 while you're using it to create. Yeah, seems 00:07:05.019 --> 00:07:07.459 like it. Yeah. It shifts processing from active 00:07:07.459 --> 00:07:09.600 discovery, active building to something more 00:07:09.600 --> 00:07:13.740 passive, like tracing or validating. Whoa. Just 00:07:13.740 --> 00:07:16.839 imagine scaling that. That reduced brain activity, 00:07:17.000 --> 00:07:19.420 that shift from active creation to passive tracing 00:07:19.420 --> 00:07:23.500 across billions of AI interactions every day 00:07:23.500 --> 00:07:25.540 all over the world. What does that mean for us, 00:07:25.600 --> 00:07:28.259 you know, long term? Yeah, it's a sobering thought. 00:07:28.399 --> 00:07:30.600 And maybe the most worrying part. These mental 00:07:30.600 --> 00:07:33.079 shortcuts, this cognitive debt, it wasn't just 00:07:33.079 --> 00:07:35.420 temporary. The effects stuck around even after 00:07:35.420 --> 00:07:37.839 they stopped using the AI. The debt lingered. 00:07:37.939 --> 00:07:40.339 Yeah. When those AI users were forced back into 00:07:40.339 --> 00:07:42.759 brain -only mode later, they consistently did 00:07:42.759 --> 00:07:44.439 worse than they did initially. It was like their 00:07:44.439 --> 00:07:46.500 mental muscles had gotten weaker from lack of 00:07:46.500 --> 00:07:50.319 use. Atrophied, kind of. That cognitive debt 00:07:50.319 --> 00:07:52.620 was real. Their ability to think independently 00:07:52.620 --> 00:07:54.860 was actually diminished. It makes me think about 00:07:54.860 --> 00:07:57.339 learning a language, you know. If you only ever 00:07:57.339 --> 00:07:59.379 use a translation app, sure, you communicate. 00:07:59.740 --> 00:08:02.199 But you're not building fluency. Take the app 00:08:02.199 --> 00:08:04.339 away, and you might actually be worse off, more 00:08:04.339 --> 00:08:06.100 dependent than when you started. You didn't build 00:08:06.100 --> 00:08:07.920 the pathways. And there was this other really 00:08:07.920 --> 00:08:11.360 fascinating, almost... artistic finding. Two 00:08:11.360 --> 00:08:14.019 English teachers graded the essays, and they 00:08:14.019 --> 00:08:16.199 could often just tell which ones were AI -assisted. 00:08:16.379 --> 00:08:18.639 Not because of mistakes, the opposite, actually. 00:08:18.699 --> 00:08:21.120 They were grammatically perfect, structured beautifully. 00:08:21.620 --> 00:08:23.500 But the teachers described them again and again 00:08:23.500 --> 00:08:26.399 as having close to perfect use of language and 00:08:26.399 --> 00:08:29.199 structure while simultaneously failing to give 00:08:29.199 --> 00:08:32.580 personal insights or clear statements. The word 00:08:32.580 --> 00:08:36.059 they kept using, soulless. That's... Quite a 00:08:36.059 --> 00:08:38.259 word. Right. That's where that homogenization 00:08:38.259 --> 00:08:40.740 fear comes in. If everyone uses the same tool 00:08:40.740 --> 00:08:44.019 to generate ideas, everyone starts sounding similar. 00:08:44.399 --> 00:08:47.059 It's this powerful force pushing towards sameness, 00:08:47.100 --> 00:08:50.059 potentially erasing unique voices, that individual 00:08:50.059 --> 00:08:53.419 spark. And honestly, I still wrestle with that 00:08:53.419 --> 00:08:55.460 myself sometimes, protecting my own voice when 00:08:55.460 --> 00:08:58.559 these tools are so easy to use. So thinking about 00:08:58.559 --> 00:09:01.639 that, if relying on AI can erode our unique creative 00:09:01.639 --> 00:09:04.460 voice, how do we actively fight against that? 00:09:04.750 --> 00:09:06.970 It has to be conscious, right? A conscious effort 00:09:06.970 --> 00:09:09.649 to cultivate your own thoughts first to protect 00:09:09.649 --> 00:09:11.950 those personal insights. Make sure your unique 00:09:11.950 --> 00:09:14.129 angle comes through. OK, so what do we do then? 00:09:14.230 --> 00:09:16.929 What's the practical takeaway? The research isn't 00:09:16.929 --> 00:09:19.529 saying AI is evil or we should ditch it. Not 00:09:19.529 --> 00:09:22.610 at all. It's more like a call for us to be much 00:09:22.610 --> 00:09:25.389 more thoughtful, much more intentional about 00:09:25.389 --> 00:09:28.279 how we use these tools. Exactly. the key idea 00:09:28.279 --> 00:09:31.120 really is understanding these two modes and choosing 00:09:31.120 --> 00:09:33.960 consciously learning mode and production mode 00:09:33.960 --> 00:09:37.519 knowing when to be in which gear right so learning 00:09:37.519 --> 00:09:39.480 mode that's when your main goal is to really 00:09:39.480 --> 00:09:42.120 get something understand a new concept build 00:09:42.120 --> 00:09:45.220 a skill here you should be brain first absolutely 00:09:45.220 --> 00:09:47.919 struggle with it yourself use ai later maybe 00:09:47.919 --> 00:09:50.179 to check your work ask questions get different 00:09:50.179 --> 00:09:52.639 explanations not to just generate the answer 00:09:52.639 --> 00:09:55.960 do the heavy lifting yourself Precisely. Ask 00:09:55.960 --> 00:09:58.620 it to explain things, not just give answers like, 00:09:58.700 --> 00:10:02.360 explain X like I'm 10, or what are three ways 00:10:02.360 --> 00:10:04.580 to think about Y? Use it like a pewter, not an 00:10:04.580 --> 00:10:07.039 answer key. And yeah, critically, protect that 00:10:07.039 --> 00:10:09.879 voice. Write the first draft yourself. Get your 00:10:09.879 --> 00:10:12.259 soul on the page first. Okay, and then separate 00:10:12.259 --> 00:10:15.340 from that is production mode. This is where AI 00:10:15.340 --> 00:10:18.220 can really help with speed, efficiency for tasks 00:10:18.220 --> 00:10:20.799 you already get, or where deep learning isn't 00:10:20.799 --> 00:10:24.299 the goal. Totally. Here, AI is great. brainstorming 00:10:24.299 --> 00:10:26.940 ideas, automating routine stuff you already know 00:10:26.940 --> 00:10:29.840 how to do, polishing your solid draft for grammar 00:10:29.840 --> 00:10:32.379 or style, summarizing long things for a quick 00:10:32.379 --> 00:10:35.440 overview. It's for execution, amplifying what 00:10:35.440 --> 00:10:38.139 you can do, not for the core learning or creating 00:10:38.139 --> 00:10:40.990 part. So the bottom line from the research seems 00:10:40.990 --> 00:10:44.070 really clear. Use AI like a cognitive crutch, 00:10:44.110 --> 00:10:46.289 letting it do the thinking for you. That can 00:10:46.289 --> 00:10:48.549 weaken your own thinking over time, but use it 00:10:48.549 --> 00:10:50.389 like a sophisticated tool you understand. That 00:10:50.389 --> 00:10:52.549 can amplify what you're already capable of. Yeah, 00:10:52.610 --> 00:10:55.409 it really just boils down to one word, intentionality. 00:10:55.610 --> 00:10:58.009 Are you consciously choosing to engage your brain 00:10:58.009 --> 00:11:00.789 to think, to learn, or are you kind of passively 00:11:00.789 --> 00:11:03.169 outsourcing that? Okay, but given that distinction. 00:11:03.799 --> 00:11:06.120 How do we make sure, like consistently, that 00:11:06.120 --> 00:11:08.000 we're picking the right mode for the right task? 00:11:08.159 --> 00:11:10.399 How do we stay disciplined about that? That's 00:11:10.399 --> 00:11:12.539 the ongoing work, isn't it? It requires constant 00:11:12.539 --> 00:11:15.940 self -awareness, pausing and asking, what's my 00:11:15.940 --> 00:11:18.820 real goal here? Learning. Or just getting it 00:11:18.820 --> 00:11:21.379 done. Purpose matters. This whole line of research, 00:11:21.480 --> 00:11:24.059 it sparks this really big, important conversation, 00:11:24.139 --> 00:11:27.889 I think, which is what's left for us? For humans, 00:11:28.090 --> 00:11:31.169 if AI can write perfectly good text, make flawless 00:11:31.169 --> 00:11:34.230 arguments, what's our unique value then? Well, 00:11:34.389 --> 00:11:36.690 those teachers who graded the essays, they kind 00:11:36.690 --> 00:11:38.509 of gave us the answer, didn't they? They valued 00:11:38.509 --> 00:11:41.049 individuality and creativity over objective perfection, 00:11:41.330 --> 00:11:44.269 that human spark, that personal insight, the 00:11:44.269 --> 00:11:46.669 wisdom from just living. That's what stood out. 00:11:47.019 --> 00:11:49.679 That's irreplaceable. AI, it's amazing, but it 00:11:49.679 --> 00:11:51.960 can't draw on a lifetime of unique experiences 00:11:51.960 --> 00:11:54.759 or real emotions or humor. It doesn't know what 00:11:54.759 --> 00:11:57.379 awe feels like or disappointment or just quiet 00:11:57.379 --> 00:11:59.039 joy. It doesn't have that embodied perspective. 00:11:59.460 --> 00:12:03.519 And that is our edge. our real competitive advantage 00:12:03.519 --> 00:12:06.299 now, our job shifts maybe. Less about generating 00:12:06.299 --> 00:12:09.220 basic info, more about infusing work with personality, 00:12:09.620 --> 00:12:12.679 telling resonant stories, connecting with other 00:12:12.679 --> 00:12:15.320 humans, bringing that soul back into the equation, 00:12:15.500 --> 00:12:18.059 even if the structure is AI -assisted sometimes. 00:12:18.379 --> 00:12:21.820 So as AI gets woven deeper into everything, the 00:12:21.820 --> 00:12:23.879 skills that matter most for us humans are going 00:12:23.879 --> 00:12:26.919 to change. Wrote memorization, less important. 00:12:27.279 --> 00:12:30.509 Generic text, that becomes a commodity. easily 00:12:30.509 --> 00:12:32.769 produced right the new essential skills are more 00:12:32.769 --> 00:12:35.629 metacognitive managing your own mind understanding 00:12:35.629 --> 00:12:38.129 how you think and learn thinking about your thinking 00:12:38.129 --> 00:12:40.929 yeah including things like tool awareness that 00:12:40.929 --> 00:12:43.970 conscious decision is this job for my brain for 00:12:43.970 --> 00:12:46.370 ai yeah or both working together making that 00:12:46.370 --> 00:12:48.590 choice deliberately and alongside that quality 00:12:48.590 --> 00:12:51.139 evaluation knowing good from bad Source verification 00:12:51.139 --> 00:12:53.460 critical now. Creative protection, guarding your 00:12:53.460 --> 00:12:56.120 unique voice. And maybe the biggest one, learning 00:12:56.120 --> 00:12:58.200 intentionality. That conscious choice to actually 00:12:58.200 --> 00:13:00.700 embrace the hard work of real learning, not just 00:13:00.700 --> 00:13:03.059 always taking the easy AI route. So if you had 00:13:03.059 --> 00:13:05.460 to pick one, what's the single most crucial skill 00:13:05.460 --> 00:13:08.440 for navigating this future that's unfolding right 00:13:08.440 --> 00:13:11.759 now? Learning intentionality, I think. And yeah, 00:13:11.899 --> 00:13:14.299 just that ability to manage your own mind. Know 00:13:14.299 --> 00:13:18.779 thyself. Applied to the age of AI. So wrapping 00:13:18.779 --> 00:13:21.820 this up. What does it all mean? Our deep dive 00:13:21.820 --> 00:13:25.159 into this MIT study, it shows AI isn't, you know, 00:13:25.159 --> 00:13:28.080 a simple savior or some kind of doom. It's way 00:13:28.080 --> 00:13:31.299 more nuanced. It's a powerful partner, yes. But 00:13:31.299 --> 00:13:34.720 crucially, a tool. not a crutch, a tool that 00:13:34.720 --> 00:13:37.600 needs skill and intention to use well. Exactly. 00:13:37.879 --> 00:13:40.139 The future isn't about rejecting these tools. 00:13:40.240 --> 00:13:42.059 That's probably impossible anyway. It's about 00:13:42.059 --> 00:13:44.700 learning to dance skillfully with them, knowing 00:13:44.700 --> 00:13:47.120 when you need to lead, when to let the AI follow 00:13:47.120 --> 00:13:49.019 your lead, and yeah, when to just step away and 00:13:49.019 --> 00:13:51.620 dance alone, trusting your own rhythm. We really 00:13:51.620 --> 00:13:53.559 encourage you listening to think about your own 00:13:53.559 --> 00:13:56.899 AI use. Are you leaning on it or are you leveraging 00:13:56.899 --> 00:13:59.179 it? Is it helping you think or thinking for you? 00:13:59.419 --> 00:14:01.940 Your brain on ChatGPT doesn't have to be a weaker 00:14:01.940 --> 00:14:03.799 brain. It can be your brain, plus this amazing 00:14:03.799 --> 00:14:06.379 tool, as long as you always, always remember 00:14:06.379 --> 00:14:08.700 who's actually supposed to be in charge. Thank 00:14:08.700 --> 00:14:11.179 you for joining us on this pretty thought -provoking 00:14:11.179 --> 00:14:13.100 deep dive. Yeah, thanks for listening. Now, if 00:14:13.100 --> 00:14:15.539 you'll excuse me, I think I might need to go 00:14:15.539 --> 00:14:17.899 write something, you know, completely unassisted, 00:14:17.940 --> 00:14:20.259 just to make sure I still can.