WEBVTT 00:00:00.000 --> 00:00:03.459 ever feel like you're only using maybe 10 % of 00:00:03.459 --> 00:00:05.740 what your AI tools can actually do. Yeah, like 00:00:05.740 --> 00:00:07.740 there's this huge potential just sitting there. 00:00:07.960 --> 00:00:10.320 Imagine unlocking this sort of artificial mind, 00:00:10.480 --> 00:00:13.039 you know? A real creative partner, ready to build 00:00:13.039 --> 00:00:16.260 something amazing with you. That's really the 00:00:16.260 --> 00:00:18.339 landscape we're diving into today. Welcome to 00:00:18.339 --> 00:00:21.780 the Deep Dive. We're all about unpacking complex 00:00:21.780 --> 00:00:24.980 ideas, making them clear, making them useful 00:00:24.980 --> 00:00:27.739 for you. And today, we're taking a good look 00:00:27.739 --> 00:00:31.480 at well, mastering chat GPT. We found this really 00:00:31.480 --> 00:00:35.100 great guide, lays out 12 pretty powerful strategies. 00:00:35.340 --> 00:00:37.960 And these aren't just, like, small tips. They're 00:00:37.960 --> 00:00:40.219 really fundamental shifts in how you interact 00:00:40.219 --> 00:00:42.939 with AI. Right. Think of it less like just a 00:00:42.939 --> 00:00:45.340 slightly smarter search engine. Yeah. And more 00:00:45.340 --> 00:00:47.579 like having this incredibly versatile collaborator. 00:00:47.780 --> 00:00:49.859 It's fascinating, isn't it, how most of us just 00:00:49.859 --> 00:00:52.299 kind of scratch the surface. We do? We treat 00:00:52.299 --> 00:00:54.299 it like, I don't know, a simple answer machine. 00:00:54.590 --> 00:00:57.210 put in a question, get an answer out. But it's 00:00:57.210 --> 00:01:01.390 so much more like a master chef. If you just 00:01:01.390 --> 00:01:04.250 walk in and say, give me some food, you'll get 00:01:04.250 --> 00:01:08.340 something. Okay, maybe edible. But generic, planned, 00:01:08.620 --> 00:01:11.719 totally. But if you give precise instructions, 00:01:12.260 --> 00:01:14.959 the ingredients, the method, the plating, suddenly 00:01:14.959 --> 00:01:17.760 you get a masterpiece. And that's the key with 00:01:17.760 --> 00:01:22.099 ChatGPT, the real power. It really lies in how 00:01:22.099 --> 00:01:25.250 we give those instructions, our prompts. That's 00:01:25.250 --> 00:01:27.450 a fantastic analogy. So, okay, let's think about 00:01:27.450 --> 00:01:29.870 that mindset then. The pro -user mindset. Yeah. 00:01:30.090 --> 00:01:31.650 They're not just taking the first answer they 00:01:31.650 --> 00:01:33.510 get. We know, not at all. It's more like a deep 00:01:33.510 --> 00:01:36.290 conversation. They engage, they refine ideas, 00:01:36.409 --> 00:01:39.189 they iterate, they actually build the concept 00:01:39.189 --> 00:01:41.890 with the AI. Exactly. It's this whole process 00:01:41.890 --> 00:01:46.200 of experimenting, learning as you go. And sometimes 00:01:46.200 --> 00:01:48.480 it messes up, but you learn from that too. Embracing 00:01:48.480 --> 00:01:51.500 the errors. OK, so moving past mindset, let's 00:01:51.500 --> 00:01:53.819 get into the real foundation, the prompts themselves. 00:01:54.060 --> 00:01:56.019 Right. And here's trick number one, which is 00:01:56.019 --> 00:01:59.060 actually pretty ingenious. Ask ChatGPT to help 00:01:59.060 --> 00:02:02.299 you write better prompts. This is so powerful. 00:02:02.439 --> 00:02:04.439 I mean, the AI can literally teach you its own 00:02:04.439 --> 00:02:06.519 language. It really can. Like, say you get a 00:02:06.519 --> 00:02:09.129 result that feels flat. A generic marketing email 00:02:09.129 --> 00:02:11.069 for a coffee shop, maybe? Yeah, just kind of, 00:02:11.069 --> 00:02:12.810 yeah. You can just ask it straight up, hey, can 00:02:12.810 --> 00:02:15.069 you improve this prompt for me? What other info 00:02:15.069 --> 00:02:16.810 should I give you to get something better? And 00:02:16.810 --> 00:02:19.330 it tells you. It'll suggest, OK, who's your target 00:02:19.330 --> 00:02:21.689 audience? What's the specific promotion? What 00:02:21.689 --> 00:02:23.789 makes your coffee shop unique? What tone do you 00:02:23.789 --> 00:02:26.050 want? A clear call to action. All the things 00:02:26.050 --> 00:02:28.849 you might have missed. Exactly. The AI kind of 00:02:28.849 --> 00:02:30.930 knows prompt engineering. It knows what it needs 00:02:30.930 --> 00:02:35.219 to do a good job. Quick tip here. Save those 00:02:35.219 --> 00:02:37.500 improved prompts. I find it helps build a library. 00:02:37.580 --> 00:02:40.500 Oh, totally you start seeing the patterns and 00:02:40.500 --> 00:02:43.960 Honestly, I still wrestle with prompt drift myself 00:02:43.960 --> 00:02:46.960 sometimes. So this trick is just it's a lifesaver 00:02:46.960 --> 00:02:49.240 really fundamental That's a huge first step. 00:02:49.360 --> 00:02:52.879 Okay moving beyond just text. What about AI's 00:02:52.879 --> 00:02:56.150 other senses? It's eyes, basically. Yeah, this 00:02:56.150 --> 00:02:58.349 is something a lot of people miss. ChatGPT's 00:02:58.349 --> 00:03:01.270 ability to analyze images, especially with GPT 00:03:01.270 --> 00:03:05.490 -4 .0, is often overlooked. Right. GPT -4, that's 00:03:05.490 --> 00:03:08.069 the latest, most all -around model, right? Does 00:03:08.069 --> 00:03:10.909 images text, audio? Precisely. It's super versatile. 00:03:11.229 --> 00:03:13.169 And it doesn't just make images. It understands 00:03:13.169 --> 00:03:16.050 them. Think about it. You snap a picture of leftovers 00:03:16.050 --> 00:03:19.169 in your fridge, some chicken, peppers, an onion, 00:03:19.169 --> 00:03:22.969 and you ask it. Give me a simple, healthy stir 00:03:22.969 --> 00:03:26.789 -fry recipe I can make in 20 minutes. Boom. Or 00:03:26.789 --> 00:03:29.150 you take a picture of some really complex chart 00:03:29.150 --> 00:03:33.009 from a report. Explain the main trends here in 00:03:33.009 --> 00:03:35.930 simple terms. Even digitizing messy handwriting. 00:03:36.229 --> 00:03:38.669 Turn it into neat bullet points. Or figure out 00:03:38.669 --> 00:03:41.069 what some weird gadget is from just a photo. 00:03:41.409 --> 00:03:43.889 The key being, you have to be specific about 00:03:43.889 --> 00:03:46.550 what you want for the image. Specificity is always 00:03:46.550 --> 00:03:49.500 key. Absolutely. So how does all this start saving 00:03:49.500 --> 00:03:52.500 us actual time in our day to day? Well, custom 00:03:52.500 --> 00:03:55.500 instructions are a massive time saver. Huge. 00:03:55.699 --> 00:03:58.580 Explain those. Basically, you tell ChatGTT up 00:03:58.580 --> 00:04:00.400 front who you are and how you want it to act. 00:04:00.479 --> 00:04:03.479 It's in the settings. Two boxes. Right. Box one. 00:04:03.840 --> 00:04:06.900 Who are you? Like, I'm a freelance content writer. 00:04:07.020 --> 00:04:09.759 I specialize in tech and startups. Focus on blog 00:04:09.759 --> 00:04:11.699 posts about new trends. Something like that. 00:04:11.719 --> 00:04:15.590 Got it. Your context. Exactly. Box two. How should 00:04:15.590 --> 00:04:19.490 chat GPT respond? Maybe professional, but approachable 00:04:19.490 --> 00:04:22.810 tone. Use clear sections, headings, bullet points, 00:04:23.050 --> 00:04:25.470 avoid jargon or academic language. And then? 00:04:26.029 --> 00:04:28.230 It just remembers that. It just remembers. No 00:04:28.230 --> 00:04:30.529 more repeating your role or preferred style in 00:04:30.529 --> 00:04:33.470 every single new chat. It just knows. Sets the 00:04:33.470 --> 00:04:36.009 stage automatically. It feels like a big step 00:04:36.009 --> 00:04:39.350 towards real personalization. Now, speaking of 00:04:39.350 --> 00:04:41.769 getting the right output. Choosing the right 00:04:41.769 --> 00:04:43.610 model is also important, isn't it? Yeah, definitely. 00:04:43.850 --> 00:04:45.589 It's like picking the right tool for the job. 00:04:45.870 --> 00:04:47.449 You wouldn't use a hammer for a screw, right? 00:04:47.670 --> 00:04:51.189 Makes sense. So GPT -40 is your great all -rounder, 00:04:51.290 --> 00:04:53.660 your default. Good for brainstorming, quick answers, 00:04:53.779 --> 00:04:55.819 that image stuff we talked about. Then you might 00:04:55.819 --> 00:04:59.199 have other options, maybe paid ones like different 00:04:59.199 --> 00:05:02.300 versions of GV4. One might be stronger for really 00:05:02.300 --> 00:05:04.519 creative stuff, emotional nuance, like writing 00:05:04.519 --> 00:05:07.500 ad copy. Right, the more creative tasks. Another 00:05:07.500 --> 00:05:10.160 might excel at complex logic, deep reasoning, 00:05:10.680 --> 00:05:13.819 maybe data analysis or coding. Needs that hardcore 00:05:13.819 --> 00:05:16.980 accuracy. So you'd pick, based on the task, social 00:05:16.980 --> 00:05:19.769 media versus a financial report. Exactly. For 00:05:19.769 --> 00:05:21.970 engaging social media, maybe that creative one, 00:05:22.269 --> 00:05:25.089 for digging into financial data or complex research, 00:05:25.550 --> 00:05:28.250 you want the one optimized for accuracy and logic. 00:05:28.810 --> 00:05:31.069 Choosing the right model really optimizes the 00:05:31.069 --> 00:05:33.790 quality you get back. Okay, so pulling this first 00:05:33.790 --> 00:05:36.949 part together, the big takeaway seems to be that 00:05:36.949 --> 00:05:40.250 really tailoring our input, our prompts, our 00:05:40.250 --> 00:05:43.250 background info, even the specific AI model we 00:05:43.250 --> 00:05:45.879 use, That's what unlocks the good stuff, right? 00:05:46.060 --> 00:05:48.579 Precisely. Specificity and context. That's how 00:05:48.579 --> 00:05:51.180 you unlock its true potential. Mid -roll sponsor 00:05:51.180 --> 00:05:55.839 raid. OK, so if you're still generating text 00:05:55.839 --> 00:05:58.240 in ChatGPT and then immediately copying it over 00:05:58.240 --> 00:06:01.000 to Google Docs or Word to actually work on it. 00:06:01.120 --> 00:06:02.819 Which, let's be honest, a lot of us do. Yeah. 00:06:03.079 --> 00:06:04.939 There might be a more efficient way right inside 00:06:04.939 --> 00:06:07.600 the tool. You mean just? editing directly in 00:06:07.600 --> 00:06:09.740 the chat window? Sort of, yeah. Think of the 00:06:09.740 --> 00:06:11.959 chat itself as your drafting space. Start with 00:06:11.959 --> 00:06:14.439 an outline product. OK. Then ask the AI to expand 00:06:14.439 --> 00:06:16.480 section two. It writes it. Then you immediately 00:06:16.480 --> 00:06:18.540 say, OK, rewrite that last paragraph, but make 00:06:18.540 --> 00:06:20.699 it more concise. Ah, so it's an iterative loop 00:06:20.699 --> 00:06:23.670 right there in the chat. Exactly. Or change the 00:06:23.670 --> 00:06:26.470 tone here, make it more persuasive. You're constantly 00:06:26.470 --> 00:06:28.870 guiding it, refining it in real time. It avoids 00:06:28.870 --> 00:06:31.269 that extra step of copy -pasting back and forth 00:06:31.269 --> 00:06:33.269 all the time. I think the flow going. Interesting. 00:06:33.430 --> 00:06:35.790 Now, what's really fascinating, building on this, 00:06:35.949 --> 00:06:40.089 is the idea of having an AI expert ready for 00:06:40.089 --> 00:06:42.970 every specific task you do. OK. And does that 00:06:42.970 --> 00:06:46.180 work? That's where custom GPTs come in. Think 00:06:46.180 --> 00:06:48.980 of them as versions of chat GPT that are already 00:06:48.980 --> 00:06:51.720 pre -trained or pre -configured for certain jobs. 00:06:51.980 --> 00:06:54.019 So instead of writing a massive prompt every 00:06:54.019 --> 00:06:56.040 time. Right. You just choose the right expert. 00:06:56.360 --> 00:06:59.360 You can browse the explore GPT section in chat 00:06:59.360 --> 00:07:01.839 GPT. You'll find things like a writing assistant 00:07:01.839 --> 00:07:04.519 or a data analyst. Maybe one specifically for 00:07:04.519 --> 00:07:07.339 Python coding. Pre -packaged expertise. Yeah. 00:07:07.579 --> 00:07:09.740 And even better, you can create your own. Imagine 00:07:09.740 --> 00:07:13.079 making, say, an SEO writing assistant just for 00:07:13.079 --> 00:07:14.839 your company. How would that work? You'd upload 00:07:14.839 --> 00:07:17.480 your specific brand guidelines, your tone of 00:07:17.480 --> 00:07:20.480 voice documents, maybe your SEO best practices 00:07:20.480 --> 00:07:22.579 checklist, keyword lists. So you feed it your 00:07:22.579 --> 00:07:27.490 rules. Exactly. Then that custom GPT... automatically 00:07:27.490 --> 00:07:30.170 writes articles that meet your standards in your 00:07:30.170 --> 00:07:33.189 voice, following your SEO rules. It's like having 00:07:33.189 --> 00:07:35.970 a specialized assistant cloned just for that 00:07:35.970 --> 00:07:38.649 task. Wow. OK, that's incredibly powerful for 00:07:38.649 --> 00:07:41.350 consistency and saving setup time. It's huge. 00:07:41.589 --> 00:07:44.410 And speaking of saving time and maintaining context, 00:07:44.829 --> 00:07:47.290 let's talk about the memory feature. Right, memory. 00:07:47.550 --> 00:07:49.089 How's that different from custom instructions? 00:07:49.370 --> 00:07:53.050 It's more granular, more specific to things you 00:07:53.050 --> 00:07:55.889 tell it during conversations. If you find yourself 00:07:55.889 --> 00:07:57.899 using the same kind of setup or persona prompt 00:07:57.899 --> 00:08:00.240 often, you can just tell it, hey, remember this. 00:08:00.300 --> 00:08:02.360 Like what? Like, remember this persona I use 00:08:02.360 --> 00:08:04.879 often. You are a marketing expert with 10 years 00:08:04.879 --> 00:08:07.920 of experience in the FMCG sector. OK. Then in 00:08:07.920 --> 00:08:09.920 later chats, you don't have to retype that. It'll 00:08:09.920 --> 00:08:12.439 just know to analyze your ideas from that specific 00:08:12.439 --> 00:08:14.920 branding perspective, maybe suggest strategies 00:08:14.920 --> 00:08:16.939 based on it. And you can manage these memories. 00:08:17.300 --> 00:08:19.519 Oh, yeah. In settings, you can see everything 00:08:19.519 --> 00:08:21.220 it's remembered. You can edit them or delete 00:08:21.220 --> 00:08:24.100 them if they're outdated. Super useful for saving 00:08:24.100 --> 00:08:26.699 specific personas, maybe formatting preferences, 00:08:27.399 --> 00:08:29.740 or details about an ongoing project you discuss 00:08:29.740 --> 00:08:32.240 frequently. Got it. Keeps things consistent without 00:08:32.240 --> 00:08:35.159 constant repetition. Now this next one sounds 00:08:35.159 --> 00:08:37.980 almost like science fiction, transforming product 00:08:37.980 --> 00:08:40.220 images. This one is pretty magical, especially 00:08:40.220 --> 00:08:42.000 if you're in marketing or run an e -commerce 00:08:42.000 --> 00:08:44.799 store. Cool. ChatGPT can literally take your 00:08:44.799 --> 00:08:47.179 product out of one photo and realistically place 00:08:47.179 --> 00:08:49.740 it into a completely new scene. Seriously? Yeah. 00:08:50.120 --> 00:08:52.580 Upload a basic product shot, let's say a simple 00:08:52.580 --> 00:08:55.299 white mug on a plain white background. OK. Standard 00:08:55.299 --> 00:08:58.139 stuff. Then you prompt it. Take this exact mug. 00:08:58.500 --> 00:09:01.000 Now, place it on a rustic oak desk, maybe next 00:09:01.000 --> 00:09:03.980 to a sleek laptop. Create an energetic morning 00:09:03.980 --> 00:09:07.200 vibe, soft sunlight streaming in, maybe even 00:09:07.200 --> 00:09:09.879 a little scheme rising from the mug. It does 00:09:09.879 --> 00:09:13.299 that. It generates an image based on that. I 00:09:13.299 --> 00:09:16.659 mean, imagine creating dozens of different lifestyle 00:09:16.659 --> 00:09:19.759 shots, ad variations, social media images, without 00:09:19.759 --> 00:09:22.399 needing a single new photo shoot. That saves. 00:09:22.789 --> 00:09:25.549 Wow, that saves a ton of money and effort. Instantly 00:09:25.549 --> 00:09:27.950 refresh your visuals. Test different backgrounds, 00:09:28.190 --> 00:09:30.409 different moods. It's incredibly powerful. OK, 00:09:30.409 --> 00:09:34.350 so these tools, the drafting in chat, custom 00:09:34.350 --> 00:09:38.429 GPTs, memory image transformation, they really 00:09:38.429 --> 00:09:41.169 embed the AI much deeper into our actual workflow, 00:09:41.289 --> 00:09:43.370 don't they? Absolutely. It stops being a separate 00:09:43.370 --> 00:09:45.389 tool you consult and starts being an integrated 00:09:45.389 --> 00:09:47.370 assistant. All right, let's shift into scaling 00:09:47.370 --> 00:09:50.289 that efficiency. Automation. You mentioned something 00:09:50.289 --> 00:09:52.750 about scheduling tasks. Yeah, this is a neat 00:09:52.750 --> 00:09:55.250 feature, especially if you have recurring information 00:09:55.250 --> 00:09:58.129 needs. With a paid account, you can essentially 00:09:58.129 --> 00:10:01.690 set up prompts that run on a schedule in spirit. 00:10:02.129 --> 00:10:04.490 What do you mean in spirit? Can it email me stuff? 00:10:04.809 --> 00:10:06.730 Not directly emailing for everyone automatically, 00:10:06.750 --> 00:10:08.769 but you can set up the generation part. Like 00:10:08.769 --> 00:10:10.929 you tell it, every weekday morning, generate 00:10:10.929 --> 00:10:13.389 a concise summary of the main U .S. stock market 00:10:13.389 --> 00:10:16.090 news from the last 24 hours, focusing on tech 00:10:16.090 --> 00:10:19.649 sector trends. Ah, OK. So it creates the content 00:10:19.649 --> 00:10:22.460 on a schedule and then I'd access it. Exactly. 00:10:22.940 --> 00:10:25.320 The idea is automating the research and compilation. 00:10:25.980 --> 00:10:28.419 Think daily industry news digests tailored to 00:10:28.419 --> 00:10:31.259 your niche, maybe weekly competitor analysis 00:10:31.259 --> 00:10:34.440 reports, or even monthly blog post ideas based 00:10:34.440 --> 00:10:37.179 on current trends. So it's like having an AI 00:10:37.179 --> 00:10:39.419 research assistant handling those repetitive 00:10:39.419 --> 00:10:41.659 info -gathering tasks. Precisely. You set it 00:10:41.659 --> 00:10:44.220 up once, and the information is ready for you 00:10:44.220 --> 00:10:46.759 when you need it. Okay, that makes sense. And 00:10:46.759 --> 00:10:49.909 what about when we're not at our desks? the mobile 00:10:49.909 --> 00:10:52.929 app experience. Oh, the mobile app has some unique 00:10:52.929 --> 00:10:55.330 tricks up its sleeve. It really leverages the 00:10:55.330 --> 00:10:57.389 phone's capabilities. Like the voice features. 00:10:57.529 --> 00:10:59.850 Yeah, the voice conversation mode is fantastic. 00:11:00.129 --> 00:11:02.309 You can genuinely practice speaking a foreign 00:11:02.309 --> 00:11:04.570 language and get real -time feedback on your 00:11:04.570 --> 00:11:06.850 pronunciation. It's pretty cool. I can see that 00:11:06.850 --> 00:11:08.950 being useful. Or just brainstorming out loud 00:11:08.950 --> 00:11:10.789 while you're walking or driving. Talk through 00:11:10.789 --> 00:11:12.750 your ideas. You can even rehearse a presentation. 00:11:12.850 --> 00:11:14.750 Just speak it to the app and ask for feedback 00:11:14.750 --> 00:11:17.330 on your pacing or clarity. And the camera integration. 00:11:17.629 --> 00:11:20.470 Super handy. for quick problem solving, snap 00:11:20.470 --> 00:11:22.870 a photo of some weird computer error message, 00:11:23.049 --> 00:11:25.590 ask, what does this mean and how do I fix it? 00:11:25.909 --> 00:11:28.529 Instant tech support. Kind of. Or translating 00:11:28.529 --> 00:11:31.490 a menu in a foreign country instantly. Just point 00:11:31.490 --> 00:11:34.289 your camera at it. The mobile app makes AI much 00:11:34.289 --> 00:11:37.190 more accessible in the moment. OK. Shifting gears 00:11:37.190 --> 00:11:40.389 a bit towards something really important. Security 00:11:40.389 --> 00:11:43.409 and privacy. What if you need to discuss sensitive 00:11:43.409 --> 00:11:46.389 stuff? Great question. That's where temporary 00:11:46.389 --> 00:11:49.299 chat mode comes in. It's crucial for anything 00:11:49.299 --> 00:11:51.919 confidential. Where do you find that? Usually 00:11:51.919 --> 00:11:54.159 in a drop down menu near the top of the chat 00:11:54.159 --> 00:11:56.539 window. When you enable it, the conversation 00:11:56.539 --> 00:11:59.059 basically becomes ephemeral. Meaning? Meaning 00:11:59.059 --> 00:12:02.419 it's not saved to your chat history. OpenAI doesn't 00:12:02.419 --> 00:12:05.039 use that data to train their models and it won't 00:12:05.039 --> 00:12:07.019 get stored in your memory feature either. So 00:12:07.019 --> 00:12:08.919 if you're talking about, say, confidential business 00:12:08.919 --> 00:12:11.759 data or personal finances. Use temporary chat. 00:12:11.919 --> 00:12:14.460 Absolutely. It's like a secure private workspace 00:12:14.460 --> 00:12:17.120 that disappears afterwards. Good to know. And 00:12:17.120 --> 00:12:19.840 finally, a seemingly simple trick, but one that 00:12:19.840 --> 00:12:22.710 probably saves a lot of future pain. organizing 00:12:22.710 --> 00:12:25.070 your chats. Oh man, this is so important. How 00:12:25.070 --> 00:12:27.009 do you find that brilliant idea you had three 00:12:27.009 --> 00:12:29.289 months ago if it's buried in hundreds of chats 00:12:29.289 --> 00:12:32.350 called new chat? Yeah, guilty is charged sometimes. 00:12:32.409 --> 00:12:35.690 It becomes a total nightmare. The key is to rename 00:12:35.690 --> 00:12:39.149 chats meaningfully right from the start. Don't 00:12:39.149 --> 00:12:41.570 just leave it as marketing email. What's a better 00:12:41.570 --> 00:12:44.129 name? Something specific like marketing ideas, 00:12:44.490 --> 00:12:48.190 Q3 2025 salted caramel coffee launch campaign. 00:12:48.470 --> 00:12:51.029 Much clearer. And remember, the search bar is 00:12:51.029 --> 00:12:53.190 pretty good. It searches titles and the content 00:12:53.190 --> 00:12:56.190 within the chats. So good naming helps immensely. 00:12:56.690 --> 00:12:59.769 Think about a system. Maybe organized by project 00:12:59.769 --> 00:13:02.990 name, by the type of task, or just by date. Anything's 00:13:02.990 --> 00:13:07.269 better than new chat 173. Right. Simple but effective 00:13:07.269 --> 00:13:10.350 hygiene. So, beyond these 12 specific tricks, 00:13:10.450 --> 00:13:13.210 the guide also touched on some broader, advanced 00:13:13.210 --> 00:13:14.950 advice. Yeah, and what's really cool here is 00:13:14.950 --> 00:13:16.909 when you start combining these tricks. Like how? 00:13:17.070 --> 00:13:19.029 Like imagine setting up custom instructions for 00:13:19.029 --> 00:13:21.230 your role, then choosing the perfect GPT model 00:13:21.230 --> 00:13:24.090 for a complex creative task, and then using that 00:13:24.090 --> 00:13:26.490 iterative drafting process right within the chat 00:13:26.490 --> 00:13:29.250 window. You layer the techniques... The synergy 00:13:29.250 --> 00:13:31.590 effect! Exactly. The whole becomes greater than 00:13:31.590 --> 00:13:33.730 the sum of its parts. The guide also stressed 00:13:33.730 --> 00:13:35.870 critical thinking, didn't it? About not just 00:13:35.870 --> 00:13:38.480 blindly trusting the output. Absolutely crucial. 00:13:38.679 --> 00:13:42.240 ChatGPT can be incredibly confident, even when 00:13:42.240 --> 00:13:44.899 it's completely wrong. These hallucinations are 00:13:44.899 --> 00:13:47.879 real. So fact checking is essential? Always, 00:13:47.879 --> 00:13:50.620 always fact check. Anything important. Financial 00:13:50.620 --> 00:13:54.500 data, medical info, key strategic points. And 00:13:54.500 --> 00:13:57.159 remember, it's a tool to support you, not replace 00:13:57.159 --> 00:14:00.480 your judgment or effort. Avoid just copy pasting 00:14:00.480 --> 00:14:04.429 that leads to plagiarism and It's about augmenting 00:14:04.429 --> 00:14:06.929 your intelligence, not outsourcing it. Perfectly 00:14:06.929 --> 00:14:10.110 put. And finally, just keep experimenting. Seriously, 00:14:10.549 --> 00:14:12.850 a tiny tweak to a prompt can sometimes lead to 00:14:12.850 --> 00:14:14.970 a dramatically better result. Don't be afraid 00:14:14.970 --> 00:14:18.129 to play around. So that final piece really boils 00:14:18.129 --> 00:14:20.629 down to using these tools thoughtfully, critically, 00:14:20.970 --> 00:14:23.350 and always verifying the output. Yes. Applying 00:14:23.350 --> 00:14:25.470 these tricks intelligently and with that critical 00:14:25.470 --> 00:14:27.950 eye, that's what really makes you an AI master. 00:14:28.169 --> 00:14:31.330 Okay, so recapping the big idea here. These 12 00:14:31.330 --> 00:14:33.649 strategies... They really form a kind of map, 00:14:33.889 --> 00:14:35.889 don't they, guiding you from maybe just dipping 00:14:35.889 --> 00:14:38.830 your toes in to really becoming proficient, a 00:14:38.830 --> 00:14:41.490 master user of AI? Yeah, it's all about saving 00:14:41.490 --> 00:14:44.350 time, definitely, but also boosting the quality 00:14:44.350 --> 00:14:46.830 of your work and honestly unlocking more of your 00:14:46.830 --> 00:14:49.669 own creative potential, too. So what's the first 00:14:49.669 --> 00:14:52.149 step someone should take after hearing all this? 00:14:52.330 --> 00:14:55.769 My advice, don't try to do all 12 at once. That's 00:14:55.769 --> 00:14:58.549 overwhelming. Yeah. Just pick one or two tricks 00:14:58.549 --> 00:15:00.669 that really jumped out at you, the ones that 00:15:00.669 --> 00:15:03.409 seem most useful for your specific work or life, 00:15:03.529 --> 00:15:06.210 and practice just those. Really go to hang of 00:15:06.210 --> 00:15:08.570 them until they feel natural, second nature. 00:15:08.990 --> 00:15:11.950 Then maybe add another one. Gradually integrate 00:15:11.950 --> 00:15:15.009 them. Makes sense. Build the habits. Right. Because 00:15:15.009 --> 00:15:17.250 the key isn't just knowing the tricks exist. 00:15:17.450 --> 00:15:20.429 It's understanding when to use which one and 00:15:20.429 --> 00:15:22.539 how to adapt them flexibly. to whatever you're 00:15:22.539 --> 00:15:24.720 trying to achieve. You know, this AI revolution, 00:15:24.840 --> 00:15:26.559 it's happening fast, these tools are here, they're 00:15:26.559 --> 00:15:29.580 powerful, beat. The real question is, how will 00:15:29.580 --> 00:15:32.860 you use them to reshape your work, your creativity, 00:15:33.320 --> 00:15:35.440 your daily tasks? A provocative thought to end 00:15:35.440 --> 00:15:38.200 on. Start experimenting today. See what you can 00:15:38.200 --> 00:15:38.600 unlock.