WEBVTT 00:00:00.000 --> 00:00:02.720 Imagine an AI that doesn't just respond, but 00:00:02.720 --> 00:00:05.759 genuinely thinks. And now, for the first time, 00:00:05.860 --> 00:00:09.460 its most advanced capabilities are really accessible 00:00:09.460 --> 00:00:12.900 to everyone. Welcome to the Deep Dive. Today, 00:00:12.939 --> 00:00:14.880 we're taking a thoughtful look at a pretty significant 00:00:14.880 --> 00:00:16.920 leap in artificial intelligence, the official 00:00:16.920 --> 00:00:20.239 launch of ChatGPT -5. This is just another update. 00:00:20.620 --> 00:00:23.120 It signals a new approach, deeply integrating 00:00:23.120 --> 00:00:27.000 reasoning and... practical problem -solving. 00:00:27.300 --> 00:00:29.359 Right. And we'll unpack the newfound simplicity 00:00:29.359 --> 00:00:32.159 of ChatGPT -5. We'll explore how its built -in 00:00:32.159 --> 00:00:34.159 reasoning actually functions and see where it 00:00:34.159 --> 00:00:36.899 stacks up against rivals like Google Gemini and 00:00:36.899 --> 00:00:40.280 Anthropic Clog 3. Then we'll dive into some honestly 00:00:40.280 --> 00:00:42.439 surprising real -world applications and think 00:00:42.439 --> 00:00:44.439 about what this evolution means for you right 00:00:44.439 --> 00:00:47.960 now. Okay, let's get into it. The most immediate 00:00:47.960 --> 00:00:50.780 change, I think, with ChatGPT -5 is its simplicity. 00:00:51.439 --> 00:00:54.159 OpenAI has pretty strategically streamlined things. 00:00:54.200 --> 00:00:56.299 They've removed all the old models. Now you've 00:00:56.299 --> 00:00:59.159 just got two clear choices. There's the default 00:00:59.159 --> 00:01:01.299 chat GPT -5, which is available to everyone, 00:01:01.840 --> 00:01:04.920 even free users, which is kind of huge. And then 00:01:04.920 --> 00:01:07.120 for those really tough problems needing more 00:01:07.120 --> 00:01:09.859 computational thinking, there's chat GPT -5 thinking. 00:01:10.159 --> 00:01:13.530 That's for the paid plans. OK, so this unified 00:01:13.530 --> 00:01:16.670 sort of hybrid model makes the user experience 00:01:16.670 --> 00:01:19.189 way simpler. No more guessing which model, no 00:01:19.189 --> 00:01:21.510 fiddling around. It just figures out your query 00:01:21.510 --> 00:01:24.090 and applies the right level of reasoning automatically. 00:01:24.629 --> 00:01:26.489 That feels like a fundamental shift, doesn't 00:01:26.489 --> 00:01:28.609 it? Absolutely. It's like, instead of digging 00:01:28.609 --> 00:01:30.569 through a toolbox for the right wrench, you have 00:01:30.569 --> 00:01:32.750 a smart toolkit that just hands it to you. It 00:01:32.750 --> 00:01:35.150 really democratizes the power, letting you focus 00:01:35.150 --> 00:01:37.810 on your problem, not the AI settings. It's a 00:01:37.810 --> 00:01:41.349 big step for mainstream use. So for most everyday 00:01:41.349 --> 00:01:43.409 stuff, does the default model basically mean 00:01:43.409 --> 00:01:45.450 you never need to think about switching? Yeah, 00:01:45.530 --> 00:01:48.310 pretty much. It aims for seamless handling, mostly 00:01:48.310 --> 00:01:50.409 removing that need for manual selection day to 00:01:50.409 --> 00:01:53.049 day. Now, what's really fascinating, maybe the 00:01:53.049 --> 00:01:55.810 biggest game changer, is this. automatic reasoning 00:01:55.810 --> 00:01:59.430 process. When you give ChatGPT 5 a complex question, 00:01:59.510 --> 00:02:02.530 you'll actually see it pause, see it think. It 00:02:02.530 --> 00:02:04.769 literally takes time, maybe 30 to 60 seconds, 00:02:05.129 --> 00:02:06.769 working through it. And the brilliant part, you 00:02:06.769 --> 00:02:08.710 get a glimpse into how it's unfolding its logic. 00:02:08.870 --> 00:02:11.490 It shows you the steps. Wow. So it's not just 00:02:11.490 --> 00:02:13.370 a fancy loading bar, then. It's actually doing 00:02:13.370 --> 00:02:15.949 something. Not at all. This built -in reasoning 00:02:15.949 --> 00:02:18.669 genuinely improves the answer quality. Think 00:02:18.669 --> 00:02:21.449 about, say, business analysis, complex problem 00:02:21.449 --> 00:02:25.590 solving, multi -step tasks. The model decomposes 00:02:25.590 --> 00:02:28.090 complex requests. That's the jargon, which just 00:02:28.090 --> 00:02:30.349 means it breaks big problems into smaller manageable 00:02:30.349 --> 00:02:32.770 steps. It tackles each part, then puts it all 00:02:32.770 --> 00:02:35.129 together into a really comprehensive, well -reasoned 00:02:35.129 --> 00:02:37.590 response. Whoa. Imagine scaling that to a billion 00:02:37.590 --> 00:02:39.810 queries. That's a lot of thinking. Seeing that 00:02:39.810 --> 00:02:41.810 logic unfold, it's like peeking behind the curtain. 00:02:42.080 --> 00:02:44.960 It really is. And that transparency, that's huge 00:02:44.960 --> 00:02:48.180 for building trust. It shifts the AI from being 00:02:48.180 --> 00:02:51.080 a black box to more of a transparent partner. 00:02:51.639 --> 00:02:54.199 When you see how it got the answer, you can understand 00:02:54.199 --> 00:02:56.219 it better, maybe even learn from its process. 00:02:56.419 --> 00:02:59.050 It helps build collaboration. How does seeing 00:02:59.050 --> 00:03:01.430 that reasoning process actually help someone 00:03:01.430 --> 00:03:04.310 trying to solve a really tricky problem? It offers 00:03:04.310 --> 00:03:07.289 crucial transparency, builds trust by revealing 00:03:07.289 --> 00:03:09.969 the AI's step -by -step logic. OK, but the whole 00:03:09.969 --> 00:03:11.990 generative AI market feels like it's moving so 00:03:11.990 --> 00:03:14.409 fast. Updates, new models popping up all the 00:03:14.409 --> 00:03:17.449 time. With all this innovation, how does ChatGPT 00:03:17.449 --> 00:03:20.870 -5 really stack up against the big rivals? That's 00:03:20.870 --> 00:03:22.569 a really important question. And yeah, there's 00:03:22.569 --> 00:03:26.069 no simple best answer. ChatGPT -5 really excels 00:03:26.069 --> 00:03:28.629 in its all -around flexibility. It's fast. Its 00:03:28.629 --> 00:03:30.849 reasoning is strong. Agent mode is effective. 00:03:31.169 --> 00:03:33.090 It's also pretty good with images, generation, 00:03:33.270 --> 00:03:35.389 and analysis. And the voice interaction feels 00:03:35.389 --> 00:03:37.870 very natural. Plus, its large context window, 00:03:37.969 --> 00:03:39.870 think of that like its working memory makes it 00:03:39.870 --> 00:03:41.810 a great general tool for businesses, developers, 00:03:42.270 --> 00:03:44.509 content creators. Right. Then you've got Google 00:03:44.509 --> 00:03:47.530 Gemini 2 .5 Pro. They are definitely leading 00:03:47.530 --> 00:03:51.310 on raw context window size. We're talking millions 00:03:51.310 --> 00:03:54.569 of tokens, the basic units of language for AI. 00:03:54.629 --> 00:03:57.490 It can handle whole books, huge code bases. So 00:03:57.490 --> 00:04:00.210 it's great for really complex logic, math problems, 00:04:00.210 --> 00:04:02.610 and that deep integration with the Google stuff 00:04:02.610 --> 00:04:05.069 drive photos. That's a big plus for some workflows. 00:04:05.409 --> 00:04:07.509 Probably ideal for researchers, data analysts, 00:04:07.689 --> 00:04:09.930 big companies dealing with tons of documents. 00:04:10.069 --> 00:04:12.330 Exactly. And we can't forget Anthropics Cl - 00:04:12.270 --> 00:04:15.710 Claude III Opus. That model is fantastic for 00:04:15.710 --> 00:04:18.290 creative writing, has this really natural human 00:04:18.290 --> 00:04:20.990 -like tone, and it's built with safety and ethics 00:04:20.990 --> 00:04:22.970 really core to its design. Its context window 00:04:22.970 --> 00:04:25.870 is very large too, bigger than GPT -4's though 00:04:25.870 --> 00:04:29.089 just behind Gemini on that specific metric. Claude 00:04:29.089 --> 00:04:31.110 III is probably the top choice for writers, marketers, 00:04:31.350 --> 00:04:33.769 anyone needing really nuanced creative communication 00:04:33.769 --> 00:04:35.990 where safety is key. This raises a really important 00:04:35.990 --> 00:04:37.810 point then. In this super competitive space, 00:04:38.129 --> 00:04:41.050 is there actually one single best model for everyone? 00:04:41.420 --> 00:04:43.759 Turns out, it's less about finding a single champion 00:04:43.759 --> 00:04:45.399 and more about finding the right tool for your 00:04:45.399 --> 00:04:47.879 specific job. Yeah. So for someone doing deep 00:04:47.879 --> 00:04:51.079 research on, say, massive legal documents, which 00:04:51.079 --> 00:04:53.399 one really stands out there? Google Gemini 2 00:04:53.399 --> 00:04:55.879 .5 Pro. Yeah. Yeah. With its market -leading 00:04:55.879 --> 00:04:58.779 context window, it's built for digesting those 00:04:58.779 --> 00:05:02.500 huge texts. Mid -roll sponsor, Red. OK. Moving 00:05:02.500 --> 00:05:04.399 beyond the big announcements, let's talk about 00:05:04.399 --> 00:05:06.500 some key features and improvements, things that 00:05:06.500 --> 00:05:09.129 actually make a difference day to day. If you've 00:05:09.129 --> 00:05:12.129 used older ChatGPT versions, I think you'll notice 00:05:12.129 --> 00:05:14.490 the writing quality improvement almost immediately. 00:05:14.949 --> 00:05:17.490 The tone feels more natural, more human, and 00:05:17.490 --> 00:05:19.930 those kind of random dashes or abrupt sentences, 00:05:20.410 --> 00:05:22.810 mostly gone. The writing just flows better. Honestly, 00:05:22.870 --> 00:05:25.310 it needs less editing afterwards. That's definitely 00:05:25.310 --> 00:05:27.149 welcome, especially if you're creating content. 00:05:27.209 --> 00:05:29.230 But what about accuracy, you know, the whole 00:05:29.230 --> 00:05:32.829 hallucination problem? Right. OpenAI says ChatGPT 00:05:32.829 --> 00:05:35.970 5 hallucinates significantly less. Now, no AI 00:05:35.970 --> 00:05:39.209 is perfect, but early signs are good. More accurate 00:05:39.209 --> 00:05:42.430 info, better source siting. I still wrestle with 00:05:42.430 --> 00:05:45.029 prompt drift myself sometimes where the AI kind 00:05:45.029 --> 00:05:46.689 of forgets what you asked for in a long chat, 00:05:46.870 --> 00:05:49.430 so fewer hallucinations is a really fundamental 00:05:49.430 --> 00:05:52.149 welcome improvement. And agent mode, which could 00:05:52.149 --> 00:05:54.790 be a bit hit or miss before, seems much more 00:05:54.790 --> 00:05:57.389 reliable now for those complex multi -step things, 00:05:57.730 --> 00:05:59.829 like researching competitors or planning strategy. 00:06:00.069 --> 00:06:02.110 How much difference does that better writing 00:06:02.110 --> 00:06:04.529 style actually make in a daily workflow? Is it 00:06:04.529 --> 00:06:06.740 really noticeable? Users are saying outputs feel 00:06:06.740 --> 00:06:09.759 way more polished, cutting down that post -generation 00:06:09.759 --> 00:06:12.439 editing time significantly. OK, let's dive into 00:06:12.439 --> 00:06:14.060 the real world stuff then, because that's where 00:06:14.060 --> 00:06:16.600 it counts. This improved agent mode sounds like 00:06:16.600 --> 00:06:19.399 it could be huge. What are some examples? For 00:06:19.399 --> 00:06:21.160 business and work, yeah, it's impressive. Take 00:06:21.160 --> 00:06:23.220 competitive analysis. You can feed it a list 00:06:23.220 --> 00:06:26.509 of competitors, ask it to analyze pricing. features, 00:06:26.769 --> 00:06:28.970 weaknesses, and it gives you back a detailed 00:06:28.970 --> 00:06:32.329 report, even a feature matrix, market gap ideas. 00:06:33.029 --> 00:06:35.430 Getting that in minutes is pretty powerful. Or 00:06:35.430 --> 00:06:37.649 for content creators, think about turning like 00:06:37.649 --> 00:06:40.430 a 45 -minute webinar into a whole package blog 00:06:40.430 --> 00:06:43.389 post, LinkedIn post, YouTube short script, even 00:06:43.389 --> 00:06:45.730 quiz questions. For marketers, it can simulate 00:06:45.730 --> 00:06:48.350 a full product launch plan. Audience analysis, 00:06:48.569 --> 00:06:51.170 keyword research, lead magnet ideas, email sequences. 00:06:51.850 --> 00:06:54.670 the works. Wow, that really is like having a 00:06:54.670 --> 00:06:56.930 whole team ready to go. What about more creative 00:06:56.930 --> 00:06:59.730 or technical tasks? Yeah, it's created drafting 00:06:59.730 --> 00:07:02.069 emails professional but friendly, like following 00:07:02.069 --> 00:07:03.829 up with a client who went quiet but focusing 00:07:03.829 --> 00:07:06.290 on adding value, not just chasing. Encoding, 00:07:06.430 --> 00:07:08.430 it built an interactive React dashboard from 00:07:08.430 --> 00:07:11.050 just a description, basically a working UI for 00:07:11.050 --> 00:07:13.730 data display. Needed some minor tweaks, but still. 00:07:14.170 --> 00:07:17.230 And for visuals, it generated complex SVG graphics, 00:07:17.569 --> 00:07:19.750 like flow charts with labels and colors, just 00:07:19.750 --> 00:07:21.810 from text. Okay, let's unpack this. agent mode 00:07:21.810 --> 00:07:23.589 a bit more. It sounds like having a research 00:07:23.589 --> 00:07:25.829 team on call almost for free. That's where it 00:07:25.829 --> 00:07:28.129 gets really interesting. The specific examples 00:07:28.129 --> 00:07:31.470 show real tangible value. There was this market 00:07:31.470 --> 00:07:34.769 research test researching organic cold brew coffee 00:07:34.769 --> 00:07:38.029 in Ho Chi Minh City. Agent mode did it in about 00:07:38.029 --> 00:07:41.209 12, 15 minutes, identified competitors, prices, 00:07:41.610 --> 00:07:44.410 distribution, marketing, trends, even suppliers. 00:07:44.810 --> 00:07:46.910 So it's not just finding stuff, it's actively 00:07:46.910 --> 00:07:48.949 pulling it together, structuring it from different 00:07:48.949 --> 00:07:51.250 places. Exactly. And the huge value is you can 00:07:51.250 --> 00:07:53.509 run these complex tasks in the background, do 00:07:53.509 --> 00:07:56.009 multiple things at once. It can search and compare 00:07:56.009 --> 00:07:59.290 across websites, navigate complex systems like 00:07:59.290 --> 00:08:01.970 research databases, filter results, and then 00:08:01.970 --> 00:08:04.470 synthesize all that data into one report with 00:08:04.470 --> 00:08:06.759 source links. Could Agent Mode handle something 00:08:06.759 --> 00:08:09.540 really complex, like summarizing quarterly financial 00:08:09.540 --> 00:08:12.000 reports from several different companies? Yes. 00:08:12.319 --> 00:08:14.800 It's explicitly designed for that synthesizing 00:08:14.800 --> 00:08:17.300 diverse data into cohesive reports from multiple 00:08:17.300 --> 00:08:20.060 sources. Now let's talk access, because this 00:08:20.060 --> 00:08:22.660 is a really key part of OpenAI's strategy here. 00:08:23.279 --> 00:08:25.879 On the free plan, you now get around 30 chat 00:08:25.879 --> 00:08:28.899 GPT -5 queries per session. Once you hit that, 00:08:29.019 --> 00:08:31.829 it switches you to GPT -5 Mini. But honestly, 00:08:32.129 --> 00:08:34.129 for simple stuff, you probably won't notice much 00:08:34.129 --> 00:08:36.409 difference in quality. And you still get access 00:08:36.409 --> 00:08:39.350 to the core features. Huh. So the absolute best 00:08:39.350 --> 00:08:41.590 model isn't totally locked behind the paywall 00:08:41.590 --> 00:08:43.309 anymore. That's actually a pretty big deal for 00:08:43.309 --> 00:08:45.129 getting more people using it, right? It's a very 00:08:45.129 --> 00:08:47.730 strategic move, making their top model accessible. 00:08:48.230 --> 00:08:50.409 Lowers the barrier to entry for advanced AI. 00:08:50.730 --> 00:08:53.769 Of course, the paid plans plus... Teams, Pro, 00:08:53.850 --> 00:08:56.529 they give you way higher limits. Dedicated access 00:08:56.529 --> 00:08:58.970 to that heavier thinking mode, priority access 00:08:58.970 --> 00:09:01.389 during busy times, and other advanced features 00:09:01.389 --> 00:09:03.909 and integrations. For a casual user hitting that 00:09:03.909 --> 00:09:06.370 free limit, is the switch to GPT -5 Mini going 00:09:06.370 --> 00:09:09.149 to feel like a big downgrade? No, probably not. 00:09:09.340 --> 00:09:12.519 For simple tasks, the quality difference is largely 00:09:12.519 --> 00:09:15.519 unnoticeable for most casual use. And thinking 00:09:15.519 --> 00:09:18.740 about integration and compatibility, Chatsheet 00:09:18.740 --> 00:09:21.240 PT5 works smoothly with all the existing features. 00:09:21.539 --> 00:09:24.100 That means image generation, canvas mode for 00:09:24.100 --> 00:09:27.159 coding and design, the deep research mode, connected 00:09:27.159 --> 00:09:29.980 apps like Google Drive. It all just works. That's 00:09:29.980 --> 00:09:32.620 smart. It's like getting all your AI Lego blocks 00:09:32.620 --> 00:09:34.940 to finally snap together perfectly without forcing 00:09:34.940 --> 00:09:37.179 them. Yeah. And if you connect that to the bigger 00:09:37.179 --> 00:09:40.490 picture, Seamless integration is just crucial 00:09:40.490 --> 00:09:43.190 for real -world adoption. AI needs to fit into 00:09:43.190 --> 00:09:45.809 your tools, not make you change everything. And 00:09:45.809 --> 00:09:47.750 looking ahead, they've announced Gmail and Google 00:09:47.750 --> 00:09:50.110 Calendar integration coming soon. I expect that 00:09:50.110 --> 00:09:52.669 will be a big boost for daily productivity. How 00:09:52.669 --> 00:09:55.809 specifically will that Gmail and Calendar integration 00:09:55.809 --> 00:09:58.090 make things better day -to -day? It'll allow 00:09:58.090 --> 00:10:00.730 direct automation. Things like drafting emails 00:10:00.730 --> 00:10:02.830 or scheduling meetings right within those apps 00:10:02.830 --> 00:10:05.149 you already use. OK, so what does all this actually 00:10:05.149 --> 00:10:07.690 mean for you, the listener? How do you get started 00:10:07.690 --> 00:10:11.009 and really use ChatGPT 5 effectively? Well, diving 00:10:11.009 --> 00:10:13.289 in is simpler than ever. Just start with tasks 00:10:13.289 --> 00:10:15.850 you already use AI for. We expect better results 00:10:15.850 --> 00:10:19.389 now. First step is easy. Just log in to ChatGPT. 00:10:19.610 --> 00:10:22.389 Version 5 is the default now. Then try different 00:10:22.389 --> 00:10:24.919 prompts. business analysis, content creation, 00:10:25.279 --> 00:10:28.080 drafting emails, analyzing data from Google Analytics. 00:10:28.500 --> 00:10:30.840 See how it handles them. And for folks on a paid 00:10:30.840 --> 00:10:33.200 plan, what's the best way to really push this 00:10:33.200 --> 00:10:35.440 agent mode? Definitely experiment with agent 00:10:35.440 --> 00:10:38.379 mode for those complex multi -step tasks. Think 00:10:38.379 --> 00:10:40.779 detailed travel planning, market research, competitive 00:10:40.779 --> 00:10:43.220 analysis, automating workflows you usually hate 00:10:43.220 --> 00:10:45.879 doing manually, and importantly, for complex 00:10:45.879 --> 00:10:48.440 problems, consciously use the reasoning feature. 00:10:48.600 --> 00:10:51.299 Let it think. Watch the indicator. Review the 00:10:51.299 --> 00:10:54.019 steps afterwards. Challenge it. Give it logic 00:10:54.019 --> 00:10:56.539 puzzles, ambiguous business scenarios. See what 00:10:56.539 --> 00:10:58.899 it can do. For someone totally new to AI, what's 00:10:58.899 --> 00:11:01.519 like the simplest but most impactful first thing 00:11:01.519 --> 00:11:03.440 they could try? Probably a straightforward content 00:11:03.440 --> 00:11:05.940 creation request, like drafting an email or summarizing 00:11:05.940 --> 00:11:08.379 a short article. Simple, but shows the power. 00:11:08.700 --> 00:11:12.100 So where does ChatGPT 5 fit into the future then? 00:11:12.259 --> 00:11:14.580 It feels like we're finally getting towards genuinely 00:11:14.580 --> 00:11:18.360 practical AI, right? Not just hype. Indeed, I 00:11:18.360 --> 00:11:20.220 think it represents a really substantial step 00:11:20.220 --> 00:11:22.919 towards practical AI assistance. The combination 00:11:22.919 --> 00:11:25.620 of reasoning, better accuracy, and those enhanced 00:11:25.620 --> 00:11:28.019 agent features make it genuinely more useful 00:11:28.019 --> 00:11:31.120 for real tasks. You'll probably notice less time 00:11:31.120 --> 00:11:34.139 editing AI stuff, more accurate research, better 00:11:34.139 --> 00:11:37.139 task completion in agent mode, more natural interactions 00:11:37.139 --> 00:11:39.100 overall. It feels like it's about augmenting 00:11:39.100 --> 00:11:41.440 our own abilities, making work more efficient, 00:11:41.659 --> 00:11:43.720 maybe more insightful, less like just a tool, 00:11:43.799 --> 00:11:46.610 more like a capable assistant. And crucially, 00:11:46.750 --> 00:11:49.350 it reduces that need to jump between different 00:11:49.350 --> 00:11:51.830 AI tools for different things. It's more unified, 00:11:51.909 --> 00:11:54.110 fits into workflows better. How quickly do you 00:11:54.110 --> 00:11:56.129 think these advanced reasoning capabilities will 00:11:56.129 --> 00:11:58.370 just become standard across all the major AI 00:11:58.370 --> 00:12:01.769 models? Competition drives that stuff fast. What 00:12:01.769 --> 00:12:04.470 fuels groundbreaking today often becomes the 00:12:04.470 --> 00:12:07.669 baseline pretty quickly tomorrow. Yeah. So to 00:12:07.669 --> 00:12:10.330 recap our deep dive today, Chad's GPT -5 marks 00:12:10.330 --> 00:12:13.389 a really pivotal moment. It's shifting advanced 00:12:13.389 --> 00:12:16.539 AI towards simplicity, built -in reasoning. and 00:12:16.539 --> 00:12:19.299 real practical use. It's designed to be genuinely 00:12:19.299 --> 00:12:22.559 helpful for your daily tasks, from complex analysis 00:12:22.559 --> 00:12:25.179 to creative work. And remember, there isn't one 00:12:25.179 --> 00:12:28.000 single best model for everything, but Chat GPT 00:12:28.000 --> 00:12:30.759 -5's agent mode and improved accuracy make it 00:12:30.759 --> 00:12:33.460 a really strong contender as an all -around tool 00:12:33.460 --> 00:12:35.379 that can definitely boost your productivity. 00:12:35.620 --> 00:12:38.120 We really encourage you to test your usual AI 00:12:38.120 --> 00:12:40.659 workflows with Chat GPT -5. See how it does. 00:12:40.899 --> 00:12:42.679 Explore new things you maybe didn't think AI 00:12:42.679 --> 00:12:44.600 could handle before, especially with that reasoning. 00:12:44.779 --> 00:12:46.539 And if you have a paid plan, definitely play 00:12:46.539 --> 00:12:48.480 with agent mode for those bigger multi -step 00:12:48.480 --> 00:12:50.840 projects. Think about updating your prompts to 00:12:50.840 --> 00:12:52.700 leverage its new strengths. And maybe longer 00:12:52.700 --> 00:12:54.539 -term, think about how this could fit deeper 00:12:54.539 --> 00:12:56.379 into your routines. Could it replace some manual 00:12:56.379 --> 00:12:59.299 research, some analysis? Stay curious, stay updated. 00:12:59.539 --> 00:13:02.179 OpenAI keeps moving fast. Here's a final thought 00:13:02.179 --> 00:13:05.200 for you to consider. ChatJPT 5 might not be the 00:13:05.200 --> 00:13:07.960 flashiest AI release we've seen lately, but its 00:13:07.960 --> 00:13:09.919 practical usefulness could actually make it the 00:13:09.919 --> 00:13:13.399 most impactful. How will you use this newly accessible 00:13:13.399 --> 00:13:16.259 power? to transform your work, your curiosity. 00:13:16.759 --> 00:13:18.580 Ultimately, it's about making AI work for you, 00:13:18.639 --> 00:13:20.480 not the other way around. Thank you for joining 00:13:20.480 --> 00:13:22.820 us on this deep dive. Out TRO music.