WEBVTT 00:00:00.000 --> 00:00:02.899 Have you ever pictured having a personal assistant 00:00:02.899 --> 00:00:05.620 who just never gets tired? Someone who's also 00:00:05.620 --> 00:00:08.000 maybe a super fast researcher, a really sharp 00:00:08.000 --> 00:00:10.560 strategist, maybe even a programmer, all rolled 00:00:10.560 --> 00:00:13.939 into one. What if I told you that picture? Well, 00:00:13.960 --> 00:00:15.779 it's becoming reality right now. Welcome to the 00:00:15.779 --> 00:00:17.820 deep dive. Today, we're not just talking about 00:00:17.820 --> 00:00:19.940 another software update. We're really digging 00:00:19.940 --> 00:00:22.739 into the arrival of Chat GPT -5. And it isn't 00:00:22.739 --> 00:00:24.980 just version 5 .0. It feels more like a quantum 00:00:24.980 --> 00:00:27.780 leap. It's starting to reshape how we work, how 00:00:27.780 --> 00:00:30.800 we create, maybe even how we think. So our mission 00:00:30.800 --> 00:00:33.820 today is to unpack what really makes Chat GPT 00:00:33.820 --> 00:00:36.549 -5. It's different, it's core advancements. And 00:00:36.549 --> 00:00:38.969 then the really interesting part, explore 10 00:00:38.969 --> 00:00:41.109 practical kind of groundbreaking applications. 00:00:41.609 --> 00:00:43.409 These are things you can use to transform workflows, 00:00:43.490 --> 00:00:45.149 whether you're in finance, content creation, 00:00:45.869 --> 00:00:47.710 even game development. We're gonna take a calm, 00:00:47.869 --> 00:00:49.770 curious look at how you can actually harness 00:00:49.770 --> 00:00:52.250 this new power. Okay, so let's unpack this a 00:00:52.250 --> 00:00:54.570 bit. We've had chat GPT before, right? What's 00:00:54.570 --> 00:00:56.229 fundamentally different this time? Why are we 00:00:56.229 --> 00:00:59.770 calling it a tectonic shift and not just an upgrade? 00:01:00.090 --> 00:01:02.009 Yeah, that's the key question. It really boils 00:01:02.009 --> 00:01:04.569 down to some fundamental changes in its architecture. 00:01:05.230 --> 00:01:07.510 The first big one is something called adaptive 00:01:07.510 --> 00:01:10.290 reasoning. Think of it like having two kind of 00:01:10.290 --> 00:01:12.430 distinct brains. There's a fast twitch one for 00:01:12.430 --> 00:01:14.469 the simple quick questions you might ask it. 00:01:14.640 --> 00:01:17.420 Bang, you get an answer. But then for the really 00:01:17.420 --> 00:01:20.000 complex stuff, the problems that need deep, multi 00:01:20.000 --> 00:01:22.760 -step thinking, it engages a deep -thought brain. 00:01:23.060 --> 00:01:25.680 And the cool part is it figures this out on its 00:01:25.680 --> 00:01:29.260 own. When you ask something, ChatGPT -5 actually 00:01:29.260 --> 00:01:31.879 assesses the complexity. It decides which brain 00:01:31.879 --> 00:01:33.959 to use and activates it seamlessly. You don't 00:01:33.959 --> 00:01:36.319 have to do anything. So you get fast answers 00:01:36.319 --> 00:01:38.799 for simple things and really thorough, well -thought 00:01:38.799 --> 00:01:41.340 -out analysis for the big strategic questions. 00:01:42.340 --> 00:01:44.879 That's a pretty big leap. It's almost like it 00:01:44.879 --> 00:01:47.459 anticipates the depth you need. And what about 00:01:47.459 --> 00:01:49.659 its memory? I know older models sometimes lost 00:01:49.659 --> 00:01:52.959 track and longer chats. Right, exactly. That 00:01:52.959 --> 00:01:56.019 leads to the second point. the vast context window. 00:01:56.260 --> 00:01:58.519 You're spot on. Older models had kind of short 00:01:58.519 --> 00:02:00.900 -term memory. They'd forget the beginning of 00:02:00.900 --> 00:02:02.640 a long conversation, which could be really frustrating, 00:02:02.900 --> 00:02:05.519 right? Led to fragmented answers, Chat GPT -5. 00:02:05.540 --> 00:02:07.540 It's got this massively expanded memory, this 00:02:07.540 --> 00:02:09.680 context window. What that means practically is 00:02:09.680 --> 00:02:13.419 you can upload, say, an entire 100 -page financial 00:02:13.419 --> 00:02:17.240 report, or a whole novel manuscript, or a complex 00:02:17.240 --> 00:02:20.000 piece of code. And it can remember, cross -reference, 00:02:20.199 --> 00:02:22.460 analyze the entire thing within one conversation, 00:02:22.560 --> 00:02:25.039 that ability to hold all the Context it's just 00:02:25.039 --> 00:02:27.360 a huge game changer for anything that needs comprehensive 00:02:27.360 --> 00:02:30.439 understanding Whoa, I mean imagine analyzing 00:02:30.439 --> 00:02:32.840 a hundred page financial report in minutes cross 00:02:32.840 --> 00:02:34.960 referencing everything instantly That's like 00:02:34.960 --> 00:02:36.979 yeah billions of queries worth of processing 00:02:36.979 --> 00:02:38.719 power right at your fingertips. That's incredible 00:02:38.719 --> 00:02:41.680 And the way it talks now it feels much more human 00:02:41.680 --> 00:02:44.900 less clunky exactly That's the third piece advanced 00:02:44.900 --> 00:02:47.520 natural language generation. It feels like it's 00:02:47.520 --> 00:02:51.689 finally crossed that uncanny valley Hmm. The 00:02:51.689 --> 00:02:53.990 communication is just remarkably human -like, 00:02:54.250 --> 00:02:57.710 much less robotic phrasing. It gets nuance. It 00:02:57.710 --> 00:03:00.550 can understand humor, specific professional tones. 00:03:00.669 --> 00:03:02.669 Like, you could ask it to write a LinkedIn post 00:03:02.669 --> 00:03:06.229 that sounds like a specific CEO or a video script 00:03:06.229 --> 00:03:08.469 in the style of a certain creator, and it nails 00:03:08.469 --> 00:03:10.930 it. It's really convincing. OK, so you put these 00:03:10.930 --> 00:03:12.710 three things together, the adaptive reasoning, 00:03:12.990 --> 00:03:16.469 the huge memory. the natural language. What's 00:03:16.469 --> 00:03:19.069 a combined effect? How does this change things 00:03:19.069 --> 00:03:21.169 beyond just getting better answers to questions? 00:03:21.409 --> 00:03:23.490 It really makes it a true cognitive partner. 00:03:23.669 --> 00:03:26.229 It amplifies our own thinking. Right. OK, now 00:03:26.229 --> 00:03:27.629 this is where it gets really interesting for 00:03:27.629 --> 00:03:29.969 everyday use. How do we take these big advancements 00:03:29.969 --> 00:03:32.310 and actually use them, turn them into tools? 00:03:32.750 --> 00:03:34.729 Let's dive into the first five applications you've 00:03:34.729 --> 00:03:38.500 mapped out. Absolutely. First up. Deep financial 00:03:38.500 --> 00:03:41.340 analysis and investment research. Imagine Chad 00:03:41.340 --> 00:03:44.680 GPT -5 as your personal round -the -clock financial 00:03:44.680 --> 00:03:47.379 analyst team. The revolution here is that traditional 00:03:47.379 --> 00:03:50.039 research plowing through 10Ks, 10Qs, earnings 00:03:50.039 --> 00:03:52.300 call transcripts and news, it takes dozens of 00:03:52.300 --> 00:03:55.960 hours, right? Grueling work. ChadJPT5 can condense 00:03:55.960 --> 00:03:58.939 that down to minutes. It's democratizing access 00:03:58.939 --> 00:04:01.319 to high -level analysis that used to be just 00:04:01.319 --> 00:04:03.199 for the pros. Okay, so it can read the reports 00:04:03.199 --> 00:04:05.419 fast, but how does it go beyond just summarizing? 00:04:05.560 --> 00:04:08.449 What makes the analysis... Yeah, good question. 00:04:08.490 --> 00:04:10.389 It's not just about reading static documents. 00:04:10.509 --> 00:04:12.810 The amazing part is its ability to access the 00:04:12.810 --> 00:04:14.909 web in real time. So you feed it the reports, 00:04:14.969 --> 00:04:16.769 you tell it your goal, sure. But then it goes 00:04:16.769 --> 00:04:18.790 out and pulls in current market news, sentiment 00:04:18.790 --> 00:04:21.529 analysis from social media, broader macroeconomic 00:04:21.529 --> 00:04:24.370 factors. It synthesizes data from dozens of live 00:04:24.370 --> 00:04:26.910 sources to give you a proper 360 degree view, 00:04:27.269 --> 00:04:29.470 something previously only big firms could really 00:04:29.470 --> 00:04:31.810 do. So what kind of output can you actually expect 00:04:31.810 --> 00:04:33.550 from this? Like what does it give you? You get 00:04:33.550 --> 00:04:35.949 structured reports, things like investment memos, 00:04:36.089 --> 00:04:38.449 SWOT analysis, even clear buy, hold, or sell 00:04:38.449 --> 00:04:41.370 recommendations with justifications. Okay, application 00:04:41.370 --> 00:04:45.850 number two, automated job and opportunity finding. 00:04:46.029 --> 00:04:48.410 This sounds useful. Yeah. Think of it as turning 00:04:48.410 --> 00:04:51.009 platforms like LinkedIn into an automated personalized 00:04:51.009 --> 00:04:54.050 opportunity engine, job hunting or even recruiting. 00:04:54.189 --> 00:04:56.170 It's often a numbers game, isn't it? So much 00:04:56.170 --> 00:04:59.209 time spent just sifting chat. GPT five can automate 00:04:59.209 --> 00:05:01.529 that sifting part. The numbers game freeing you 00:05:01.529 --> 00:05:04.569 up to focus on the quality side, building relationships, 00:05:04.769 --> 00:05:06.389 prepping for interviews, making connections. 00:05:06.509 --> 00:05:09.050 How does it actually do that? And are there security 00:05:09.050 --> 00:05:11.850 things to watch out for? Right. So you enable 00:05:11.850 --> 00:05:14.240 something called agent mode. That basically lets 00:05:14.240 --> 00:05:16.720 the AI take actions for you in a web browser. 00:05:17.160 --> 00:05:19.379 clicking links, filling forms. Then you give 00:05:19.379 --> 00:05:21.660 it really detailed criteria, what jobs you want, 00:05:21.740 --> 00:05:24.220 what skills, location, company size, et cetera. 00:05:24.699 --> 00:05:27.199 Now, security is crucial here. If you give it 00:05:27.199 --> 00:05:30.060 login info, be super careful. Maybe use a temporary 00:05:30.060 --> 00:05:32.980 password, change it right after, and never share 00:05:32.980 --> 00:05:35.120 two -factor authentication codes. Seriously. 00:05:35.639 --> 00:05:37.959 But with that caution, yeah, it can browse job 00:05:37.959 --> 00:05:40.240 boards or candidate profiles, filter based on 00:05:40.240 --> 00:05:42.660 your needs, and give you a short list. For job 00:05:42.660 --> 00:05:45.199 seekers, it's top opportunities. For hiring managers, 00:05:45.379 --> 00:05:47.779 it's potential candidates. often with analysis 00:05:47.779 --> 00:05:49.860 included. How specific can you get with those 00:05:49.860 --> 00:05:52.199 criteria for it to be really effective? Very 00:05:52.199 --> 00:05:54.980 specific, down to years of experience, specific 00:05:54.980 --> 00:05:57.399 skills, company type, even things like open source 00:05:57.399 --> 00:05:59.240 contributions if you're looking for developers. 00:05:59.740 --> 00:06:02.300 Interesting. Okay, number three, content strategy 00:06:02.300 --> 00:06:06.430 and video ideation. Big one for creators. Huge. 00:06:06.750 --> 00:06:08.850 This is like having a content strategist who 00:06:08.850 --> 00:06:10.949 analyzes your competitors and your own channel 00:06:10.949 --> 00:06:14.009 to brainstorm video ideas. Normally, this is 00:06:14.009 --> 00:06:16.589 super time consuming, right? Watching hours of 00:06:16.589 --> 00:06:19.629 competitor videos, taking notes, chat GPT -5 00:06:19.629 --> 00:06:21.829 can look across multiple channels simultaneously, 00:06:22.189 --> 00:06:24.449 spotting patterns, successful formats, title, 00:06:24.569 --> 00:06:27.050 styles, things a human might easily miss. Okay, 00:06:27.149 --> 00:06:29.759 but does it just copy? what's already popular. 00:06:30.139 --> 00:06:32.360 Or can it actually generate new ideas? It really 00:06:32.360 --> 00:06:34.360 synthesizes creatively. It doesn't just copy. 00:06:34.540 --> 00:06:37.379 You give it, say, links to five, 10 creators 00:06:37.379 --> 00:06:39.579 or blogs you like, plus your own channel. Ask 00:06:39.579 --> 00:06:42.040 it to identify what works for them, themes, formats, 00:06:42.379 --> 00:06:45.319 titles. Then it generates maybe 10 new unique 00:06:45.319 --> 00:06:48.180 ideas that blend those best practices with what's 00:06:48.180 --> 00:06:50.120 already worked on your channel. It finds fresh 00:06:50.120 --> 00:06:53.339 angles. Application four, website and profile 00:06:53.339 --> 00:06:55.519 optimization. like getting instant feedback. 00:06:55.980 --> 00:06:58.560 Exactly. Think of it as having a UX UI consultant 00:06:58.560 --> 00:07:00.980 and a copywriter on call instantly. This is a 00:07:00.980 --> 00:07:03.300 big deal because hiring consultants is expensive, 00:07:03.420 --> 00:07:06.120 right? Chat GPT -5, especially with its new image 00:07:06.120 --> 00:07:08.579 analysis, can look at a screenshot of your website 00:07:08.579 --> 00:07:11.079 or profile and give you valuable feedback, pointing 00:07:11.079 --> 00:07:13.379 out weaknesses you might be blind to. Can it 00:07:13.379 --> 00:07:16.339 go beyond just suggesting visual changes? Can 00:07:16.339 --> 00:07:18.779 it help improve actual engagement, getting people 00:07:18.779 --> 00:07:21.209 to click or sign up? Yeah, definitely. It looks 00:07:21.209 --> 00:07:23.850 deeper than just visuals. You upload a good, 00:07:24.069 --> 00:07:26.250 clear screenshot of your home page, a landing 00:07:26.250 --> 00:07:28.970 page, your LinkedIn profile header. Tell it the 00:07:28.970 --> 00:07:30.790 goal of that page. What do you want people to 00:07:30.790 --> 00:07:33.490 do there? Then ask for specific feedback on the 00:07:33.490 --> 00:07:35.410 headline, the call -to -action button, the layout. 00:07:35.870 --> 00:07:38.089 It analyzes the copywriting, the design, the 00:07:38.089 --> 00:07:41.290 user experience UX, and conversion optimization, 00:07:41.589 --> 00:07:44.310 CRO. It'll point out distractions or friction 00:07:44.310 --> 00:07:46.269 points that stop users from taking that desired 00:07:46.269 --> 00:07:49.750 action and suggest fixes. Number five. better 00:07:49.750 --> 00:07:53.050 content creation in your voice. This echoes a 00:07:53.050 --> 00:07:54.850 common AI complaint, right? That generic sound. 00:07:55.170 --> 00:08:22.980 Yes, the lack of authenticity. So, does that 00:08:22.980 --> 00:08:25.980 mean it can really... write exactly like me, 00:08:26.060 --> 00:08:27.680 even capture, I don't know, my specific sense 00:08:27.680 --> 00:08:29.519 of humor or the way I freeze things. It gets 00:08:29.519 --> 00:08:32.080 remarkably close. It brings your tone, nuance, 00:08:32.220 --> 00:08:34.580 word choices, sentence structure, even humor, 00:08:34.740 --> 00:08:37.799 yeah, for really authentic content. Setting it 00:08:37.799 --> 00:08:40.720 up involves creating a custom GPT in the interface. 00:08:41.399 --> 00:08:44.340 Then you gather maybe 10, 20 pieces of your own 00:08:44.340 --> 00:08:46.960 writing blog posts, newsletters, whatever represents 00:08:46.960 --> 00:08:49.820 your voice well. You upload those to its knowledge 00:08:49.820 --> 00:08:52.019 section, and the key step is instructing it. 00:08:52.790 --> 00:08:55.350 Analyze these documents. Learn my style. Pay 00:08:55.350 --> 00:08:57.590 attention to sentence length, word choice, tone, 00:08:57.889 --> 00:09:00.549 jargon, how I build arguments. Then you can ask 00:09:00.549 --> 00:09:02.710 it to draft LinkedIn posts, Instagram scripts, 00:09:02.929 --> 00:09:05.389 newsletters, all sounding like you. You know, 00:09:05.409 --> 00:09:07.309 I'll admit, I still wrestle with prompt drifts 00:09:07.309 --> 00:09:09.750 sometimes myself. The AI can kind of wander off 00:09:09.750 --> 00:09:11.970 style if you're not precise. But training a custom 00:09:11.970 --> 00:09:14.190 GPT on my own voice, it really does help keep 00:09:14.190 --> 00:09:17.840 things consistent. They really show the AI acting 00:09:17.840 --> 00:09:19.639 almost like a personalized expert in different 00:09:19.639 --> 00:09:21.980 fields. What's the main takeaway here? They free 00:09:21.980 --> 00:09:23.960 us up from the repetitive stuff, letting us focus 00:09:23.960 --> 00:09:26.399 more on the creative, strategic thinking. Mid 00:09:26.399 --> 00:09:28.639 -roll sponsor read. Welcome back to The Deep 00:09:28.639 --> 00:09:31.240 Dive. We've looked at how ChatGPT -5 can change 00:09:31.240 --> 00:09:34.039 finance, job searching content. Now let's explore 00:09:34.039 --> 00:09:36.759 five more really powerful ways you can use this. 00:09:37.019 --> 00:09:39.669 This cognitive partner. Okay, next up, number 00:09:39.669 --> 00:09:42.750 six, advanced trip and event planning. Think 00:09:42.750 --> 00:09:44.830 of this as your super -powered personal travel 00:09:44.830 --> 00:09:47.789 agent and event planner combined. The breakthrough 00:09:47.789 --> 00:09:50.350 is that planning a trip, even a simple one, involves 00:09:50.350 --> 00:09:53.269 juggling so many variables, flights, hotels, 00:09:53.789 --> 00:09:56.870 activities, budget, logistics. Chat GPT -5 can 00:09:56.870 --> 00:09:59.440 handle all of that simultaneously. It searches 00:09:59.440 --> 00:10:01.620 the web for real -time prices and availability, 00:10:02.340 --> 00:10:04.559 considers your preferences, and spits out comprehensive 00:10:04.559 --> 00:10:07.279 plans in minutes. In that deep thought capability 00:10:07.279 --> 00:10:09.700 we mentioned, it uses that to factor in things 00:10:09.700 --> 00:10:12.240 like travel time between locations, museum opening 00:10:12.240 --> 00:10:14.360 hours, even local festivals happening during 00:10:14.360 --> 00:10:16.490 your stay. That sounds useful. But what about 00:10:16.490 --> 00:10:18.769 curveballs? Can it handle unexpected changes, 00:10:18.850 --> 00:10:21.129 like if it suddenly rains on your planned beach 00:10:21.129 --> 00:10:23.090 day? It's actually pretty adaptable. You give 00:10:23.090 --> 00:10:25.830 it the basics, an estimation, dates, budget, 00:10:25.929 --> 00:10:28.409 who's going. But the key is being specific about 00:10:28.409 --> 00:10:32.289 interests. Are you a foodie? History nerd. Looking 00:10:32.289 --> 00:10:34.789 for business contacts. Tell it. The more detail, 00:10:35.029 --> 00:10:37.529 the better the plan. And yes, you can absolutely 00:10:37.529 --> 00:10:40.029 ask for alternatives. Give me a Plan B for rainy 00:10:40.029 --> 00:10:42.370 days, for example. You'll get detailed daily 00:10:42.370 --> 00:10:44.750 itineraries, recommendations tailored to you, 00:10:44.889 --> 00:10:47.149 maybe even potential business meetups if you 00:10:47.149 --> 00:10:49.870 ask for that, and a budget breakdown. Yeah, it 00:10:49.870 --> 00:10:52.629 can build in backup options. Okay, number seven. 00:10:53.029 --> 00:10:56.519 2D game creation for engagement. This one sounds 00:10:56.519 --> 00:10:59.240 kind of wild. Building games with no code. It 00:10:59.240 --> 00:11:01.460 is pretty wild. The revolution is that game development, 00:11:01.539 --> 00:11:03.919 even for simple 2D games, always needed programming 00:11:03.919 --> 00:11:07.779 skills, right? HTML, CSS, JavaScript. Now there's 00:11:07.779 --> 00:11:10.259 this thing called Canvas Mode. It's like an interactive 00:11:10.259 --> 00:11:13.080 space where ChatGPT 5 can generate actual runnable 00:11:13.080 --> 00:11:15.620 applications based on your description. You literally 00:11:15.620 --> 00:11:17.220 describe the game you want in plain English, 00:11:17.279 --> 00:11:20.299 and it builds it. This opens up crazy new possibilities 00:11:20.299 --> 00:11:22.539 for marketing, education, internal training. 00:11:22.659 --> 00:11:24.340 Wait, so a complete beginner, someone who's never 00:11:24.340 --> 00:11:26.559 coded could just describe a game and it builds 00:11:26.559 --> 00:11:29.259 it. Absolutely, that's the idea. You enable canvas 00:11:29.259 --> 00:11:32.559 mode, you get specific, create a top -down space 00:11:32.559 --> 00:11:34.899 shooter, player ship moves with arrow keys, fires 00:11:34.899 --> 00:11:36.860 with space bar, asteroids come from the top, 00:11:37.200 --> 00:11:40.279 it generates the HTML, CSS, JavaScript, you can 00:11:40.279 --> 00:11:42.259 preview it right there. Then you can say, okay. 00:11:42.360 --> 00:11:45.159 Make the ship faster or add a scoring system. 00:11:45.279 --> 00:11:47.080 Think about using it for like an interactive 00:11:47.080 --> 00:11:49.620 explainer for compound interest or a little tool 00:11:49.620 --> 00:11:52.360 to let people simulate decorating a room for 00:11:52.360 --> 00:11:55.580 a furniture store or gamified quizzes for marketing. 00:11:55.720 --> 00:11:59.120 Okay, number eight. Internal tools and dashboard 00:11:59.120 --> 00:12:01.899 creation. Building custom software interfaces. 00:12:02.279 --> 00:12:04.759 Yeah, specifically the front ends, what the user 00:12:04.759 --> 00:12:06.980 sees and interacts with. The pain point this 00:12:06.980 --> 00:12:09.460 solves is that every business has unique workflows, 00:12:09.820 --> 00:12:11.980 right? And off -the -shelf software rarely fits 00:12:11.980 --> 00:12:14.539 perfectly. But getting custom tools built is 00:12:14.539 --> 00:12:17.200 usually super expensive and takes ages. Chat 00:12:17.200 --> 00:12:20.240 GPT -5 lets you rapidly prototype and even build 00:12:20.240 --> 00:12:22.539 functional front ends for these internal tools, 00:12:22.860 --> 00:12:24.919 streamlining things without that huge development 00:12:24.919 --> 00:12:26.980 cost. So it's more about speeding up the process 00:12:26.980 --> 00:12:29.299 and maybe building simpler tools, not necessarily 00:12:29.299 --> 00:12:31.240 replacing a full development team for complex 00:12:31.240 --> 00:12:33.799 stuff? Exactly. It's about faster prototyping 00:12:33.799 --> 00:12:35.259 and getting functional front ends up quickly 00:12:35.259 --> 00:12:37.980 to streamline operations. You describe the workflow. 00:12:38.240 --> 00:12:40.379 Who uses it? What do they need to do? What data 00:12:40.379 --> 00:12:43.039 should be shown? Specify the UI bits you need. 00:12:43.080 --> 00:12:45.679 Tables, forms, buttons, dropdowns. It generates 00:12:45.679 --> 00:12:48.419 the code. You test, refine. You could build, 00:12:48.539 --> 00:12:51.039 say, a simple dashboard to manage blog content, 00:12:51.580 --> 00:12:54.820 a basic CRM to track leads, or a customer calculator 00:12:54.820 --> 00:12:59.580 for a specific business need. 9. Custom AI Advisors 00:12:59.580 --> 00:13:02.870 and Mentors That sounds intriguing, like building 00:13:02.870 --> 00:13:05.350 your own virtual advisors. Yeah, it's fascinating. 00:13:05.429 --> 00:13:07.789 The idea is to build personalized advice systems 00:13:07.789 --> 00:13:10.269 based on the thinking of leaders or experts you 00:13:10.269 --> 00:13:12.509 really admire. The revolution is, you know, most 00:13:12.509 --> 00:13:14.909 of us can't just call up Ray Dalio or Brene Brown 00:13:14.909 --> 00:13:16.950 for advice, right? But their knowledge books, 00:13:17.070 --> 00:13:19.950 interviews, talks, it's out there. Chat GPT -5 00:13:19.950 --> 00:13:21.970 can ingest all that public knowledge. You distill 00:13:21.970 --> 00:13:25.090 it into a custom GPT, creating the sort of virtual 00:13:25.090 --> 00:13:27.350 boardroom of mentors you can consult anytime. 00:13:27.610 --> 00:13:30.190 OK, but how reliable is that? Are these virtual 00:13:30.190 --> 00:13:31.980 mentors? towards really capturing the essence 00:13:31.980 --> 00:13:34.539 or just spinning back keywords, how does it compare 00:13:34.539 --> 00:13:37.080 to real advice? Well, they're synthesizing the 00:13:37.080 --> 00:13:39.759 documented knowledge, so they offer perspectives 00:13:39.759 --> 00:13:42.059 consistent with that person's state of philosophies. 00:13:42.340 --> 00:13:45.480 It's not new insight, but it's diverse structured 00:13:45.480 --> 00:13:48.639 thinking based on their work. To set it up, you'd 00:13:48.639 --> 00:13:51.080 pick maybe three to five diverse thinkers. Ask 00:13:51.080 --> 00:13:53.899 ChatGPT5 to research them deeply, create summary 00:13:53.899 --> 00:13:56.720 documents of their core ideas, upload those docs 00:13:56.720 --> 00:14:00.899 to a new custom GPT's knowledge base. Act as 00:14:00.899 --> 00:14:03.399 an advisory board. When I ask a question, answer 00:14:03.399 --> 00:14:05.240 from each perspective, grounding it in their 00:14:05.240 --> 00:14:08.700 philosophy. Using voice mode could make it feel 00:14:08.700 --> 00:14:11.039 quite immersive, like a real discussion. You 00:14:11.039 --> 00:14:12.820 could ask about tough hiring choices, getting 00:14:12.820 --> 00:14:14.820 perspectives from different angles, or how to 00:14:14.820 --> 00:14:17.059 approach innovation based on someone like Naval 00:14:17.059 --> 00:14:19.659 Ravikant's ideas. And the final one, number 10, 00:14:20.159 --> 00:14:22.960 AI agent creation. and workflow automation. This 00:14:22.960 --> 00:14:24.639 sounds like the most advanced one. It definitely 00:14:24.639 --> 00:14:27.179 builds on the previous ideas. This is about creating 00:14:27.179 --> 00:14:30.340 complex automation systems where ChatGPT5 is 00:14:30.340 --> 00:14:32.720 the brain making decisions within the workflow. 00:14:33.179 --> 00:14:35.960 The huge leap here. Previous tools like Zepier 00:14:35.960 --> 00:14:38.159 or Make were great for connecting apps. If this 00:14:38.159 --> 00:14:41.799 happens, then do that. Simple triggers. But ChatGPT5 00:14:41.799 --> 00:14:44.419 adds reasoning. The workflow can now say, OK, 00:14:44.460 --> 00:14:46.580 this email came in, let me think about its content, 00:14:46.960 --> 00:14:49.139 maybe research the sender on LinkedIn, and then 00:14:49.139 --> 00:14:51.379 decide whether to draft a reply. forward it, 00:14:51.440 --> 00:14:53.820 or create a task. It's that decision -making 00:14:53.820 --> 00:14:55.919 layer based on understanding the context that's 00:14:55.919 --> 00:14:58.960 new. You can use platforms like Make .com, N8n 00:14:58.960 --> 00:15:01.860 .io, or Lindy AI, which integrate this kind of 00:15:01.860 --> 00:15:03.860 reasoning. So is setting these agents up something 00:15:03.860 --> 00:15:06.620 only coders can do, or is it becoming more accessible? 00:15:06.960 --> 00:15:08.620 It's becoming more accessible, especially with 00:15:08.620 --> 00:15:11.500 visual platforms like Make .com. While the potential 00:15:11.500 --> 00:15:14.039 is really powerful, you don't necessarily need 00:15:14.039 --> 00:15:17.419 to be a hardcore developer. Think about intelligent 00:15:17.419 --> 00:15:20.309 content automation. An RSS feed publishes an 00:15:20.309 --> 00:15:23.370 article. The agent grabs it, sends it to ChatGPT5 00:15:23.370 --> 00:15:26.090 to summarize and pull key points, then drafts 00:15:26.090 --> 00:15:28.750 a LinkedIn post in your trained voice, maybe 00:15:28.750 --> 00:15:31.850 even suggests relevant hashtags, or optimized 00:15:31.850 --> 00:15:34.429 daily planning. Pulls your calendar and task 00:15:34.429 --> 00:15:37.549 list each morning, sends it to ChatGPT5 to create 00:15:37.549 --> 00:15:39.769 an hour -by -hour schedule optimized for focus, 00:15:40.049 --> 00:15:42.450 batches similar tasks, and sends it to your Slack. 00:15:42.620 --> 00:15:45.860 These last five really push into AI making decisions 00:15:45.860 --> 00:15:48.220 and acting more autonomously. What's the big 00:15:48.220 --> 00:15:50.720 takeaway from this set? It's moving towards AI 00:15:50.720 --> 00:15:53.220 as a genuine collaborator, automating not just 00:15:53.220 --> 00:15:55.740 tasks, but complex decision -making processes, 00:15:56.179 --> 00:15:57.960 pushing our potential. So let's pull this all 00:15:57.960 --> 00:16:00.019 together. We've covered a lot of ground. What's 00:16:00.019 --> 00:16:02.220 the big idea we should walk away with? It seems 00:16:02.220 --> 00:16:04.559 clear. ChatGPT 5 isn't just a better chat bot. 00:16:04.720 --> 00:16:06.480 It's something more fundamental. It's acting 00:16:06.480 --> 00:16:08.799 like a cognitive partner, genuinely boosting 00:16:08.799 --> 00:16:11.179 our productivity and how we make decisions in 00:16:11.440 --> 00:16:14.080 Well, almost anything we do. Absolutely. And 00:16:14.080 --> 00:16:16.240 I think the key takeaway for succeeding in this 00:16:16.240 --> 00:16:19.279 new era is learning to collaborate with AI, not 00:16:19.279 --> 00:16:21.139 trying to compete against it. It's about partnership. 00:16:21.360 --> 00:16:24.360 Right. So the action plan for everyone listening, 00:16:24.539 --> 00:16:27.759 don't wait. Start experimenting now. Maybe pick 00:16:27.759 --> 00:16:29.860 one thing we talked about today that resonates 00:16:29.860 --> 00:16:32.179 with a challenge you have. Try the financial 00:16:32.179 --> 00:16:35.659 analysis on a stock. Set up that custom GPT for 00:16:35.659 --> 00:16:38.460 your writing voice. Try building a simple dashboard 00:16:38.460 --> 00:16:41.100 for that annoying task you do every week. And 00:16:41.100 --> 00:16:43.960 remember those tips for success. Be super specific 00:16:43.960 --> 00:16:46.299 with your instructions, your prompts. If you 00:16:46.299 --> 00:16:49.360 use agent mode, watch it closely. Supervise it, 00:16:49.679 --> 00:16:52.059 and always double check the outputs. Use it as 00:16:52.059 --> 00:16:54.240 an amazing research assistant, not an infallible 00:16:54.240 --> 00:16:56.759 oracle. And of course, the security stuff. Guard 00:16:56.759 --> 00:16:59.360 your logins carefully, avoid putting super sensitive 00:16:59.360 --> 00:17:01.559 company or personal data into it, and just be 00:17:01.559 --> 00:17:04.059 aware that AI can sometimes have biases based 00:17:04.059 --> 00:17:06.299 on its training data. Yeah, the future of work, 00:17:06.519 --> 00:17:09.579 it really is here, and it's powered by AI. The 00:17:09.579 --> 00:17:11.500 folks who learn to use these tools effectively 00:17:11.500 --> 00:17:13.799 today are absolutely going to be the ones leading 00:17:13.799 --> 00:17:16.700 tomorrow. So the question for you listening is, 00:17:17.160 --> 00:17:19.319 which one of these use cases are you most excited 00:17:19.319 --> 00:17:21.559 to try first? What's your biggest challenge that 00:17:21.559 --> 00:17:24.680 maybe, just maybe, this could help solve? We 00:17:24.680 --> 00:17:27.099 really hope this deep dive gave you a solid map 00:17:27.099 --> 00:17:29.680 for starting to harness the power of ChatGP T5. 00:17:29.900 --> 00:17:32.240 Until next time, keep exploring. Out to your 00:17:32.240 --> 00:17:32.599 own music.