WEBVTT 00:00:00.000 --> 00:00:03.319 Imagine getting a hospital bill for $195 ,000. 00:00:03.740 --> 00:00:07.419 It's overwhelming, debilitating, really, especially 00:00:07.419 --> 00:00:10.140 coming right after a family tragedy. But what 00:00:10.140 --> 00:00:12.460 if a simple AI chatbot, something you could just 00:00:12.460 --> 00:00:14.880 type into, could help you cut that crushing expense 00:00:14.880 --> 00:00:19.100 down, all the way down to just $33 ,000? That's 00:00:19.100 --> 00:00:21.160 not some future tech. It's financial justice 00:00:21.160 --> 00:00:24.160 happening now, powered by AI like Claude. And 00:00:24.160 --> 00:00:25.859 it really changes how we think about consumer 00:00:25.859 --> 00:00:28.579 protection. Welcome to this deep dive. We take 00:00:28.579 --> 00:00:31.000 complex tech sources, all the latest intelligence, 00:00:31.239 --> 00:00:33.600 and turn it into clear, actionable insights just 00:00:33.600 --> 00:00:35.859 for you. Today, we're going to unpack three really 00:00:35.859 --> 00:00:38.460 critical threads. First up, that incredible story 00:00:38.460 --> 00:00:41.520 using AI to fight back against massive medical 00:00:41.520 --> 00:00:43.859 building issues. Fraud, really. Then we need 00:00:43.859 --> 00:00:45.780 to look beyond the hype. Everyone talks LLMs, 00:00:45.840 --> 00:00:48.200 but we'll define the eight essential AI model 00:00:48.200 --> 00:00:50.240 types you actually need to understand for 2025. 00:00:50.759 --> 00:00:53.939 And finally, the money behind it all. The staggering 00:00:53.939 --> 00:00:56.659 financial reality of the AI boom. the company 00:00:56.659 --> 00:00:59.359 that just blew past $5 trillion in value. Okay, 00:00:59.399 --> 00:01:01.899 let's unpack this. So this story, it came from 00:01:01.899 --> 00:01:04.239 a user on Threads, and it's just a perfect example 00:01:04.239 --> 00:01:07.879 of AI as a, well, a counterweapon against confusing 00:01:07.879 --> 00:01:10.739 systems. After a family member tragically passed 00:01:10.739 --> 00:01:13.219 away only four hours in the hospital, and crucially, 00:01:13.340 --> 00:01:16.980 their insurance had lapsed, the bill came $195 00:01:16.980 --> 00:01:20.840 ,000. Wow. Almost $200 ,000 for just four hours. 00:01:20.900 --> 00:01:24.409 That figure. It's only possible because, frankly, 00:01:24.530 --> 00:01:26.689 hospitals often bank on complexity. They overwhelm 00:01:26.689 --> 00:01:28.829 patients. But the user took a critical first 00:01:28.829 --> 00:01:31.250 step. They insisted the hospital provide itemized 00:01:31.250 --> 00:01:34.390 CPT codes. That was key. Right. And just so everyone's 00:01:34.390 --> 00:01:36.530 clear, CPT codes, current procedural terminology 00:01:36.530 --> 00:01:38.650 codes, they're these standardized five -digit 00:01:38.650 --> 00:01:40.870 medical codes. They describe every single procedure, 00:01:41.150 --> 00:01:43.510 service, even devices used. And that's often 00:01:43.510 --> 00:01:46.150 where the sneaky duplication and potential fraud 00:01:46.150 --> 00:01:49.709 hides, in those details. Exactly. So the user 00:01:49.709 --> 00:01:52.469 fed those codes into an AI, Claw in this case, 00:01:52.590 --> 00:01:55.409 and the AI acted like a super specialized auditor. 00:01:56.230 --> 00:01:58.609 Immediately, it started flagging specific violations, 00:01:58.950 --> 00:02:01.329 loads of them, the biggest one. Billing for a 00:02:01.329 --> 00:02:03.530 master procedure, the main thing they did, and 00:02:03.530 --> 00:02:05.569 then separately billing again for all the smaller 00:02:05.569 --> 00:02:08.050 things included within it, that duplication alone. 00:02:08.189 --> 00:02:11.939 It accounted for over $100 ,000 of, well... BS 00:02:11.939 --> 00:02:14.360 charges. That's just a fundamental violation, 00:02:14.620 --> 00:02:17.580 right, of standard medical coding, a huge Medicare 00:02:17.580 --> 00:02:19.840 no -no. Even if Medicare wasn't directly involved 00:02:19.840 --> 00:02:22.560 here, it sets the standard. Totally. And Claude 00:02:22.560 --> 00:02:24.759 didn't stop there. It spotted charges for services 00:02:24.759 --> 00:02:27.780 labeled inpatient only, but the patient was never 00:02:27.780 --> 00:02:29.780 formally admitted. They were only in the ER for 00:02:29.780 --> 00:02:32.060 four hours. And another big systemic error it 00:02:32.060 --> 00:02:34.580 found, charging for both ventilator use and critical 00:02:34.580 --> 00:02:37.020 care on the same day. Medicare explicitly says 00:02:37.020 --> 00:02:38.990 you can't do that. And it wasn't just the procedures. 00:02:39.110 --> 00:02:42.210 The markup on basic supplies was, well, staggering. 00:02:42.569 --> 00:02:44.789 Claude found the hospital charging anywhere from 00:02:44.789 --> 00:02:47.889 500 percent up to 2300 percent above the standard 00:02:47.889 --> 00:02:50.729 rates for things like gauze pads, IV supplies, 00:02:50.990 --> 00:02:54.270 basic stuff. It's clearly a strategy. You know, 00:02:54.310 --> 00:02:56.430 the hospital assumes the patient will be too 00:02:56.430 --> 00:02:59.930 overwhelmed by grief or just too uninformed to 00:02:59.930 --> 00:03:01.810 fight back on these technical codes and percentages. 00:03:02.069 --> 00:03:05.250 So the user took Claude's findings, used the 00:03:05.250 --> 00:03:08.030 AI to help write a really detailed letter citing 00:03:08.030 --> 00:03:10.669 every single violation using the right kind of... 00:03:11.299 --> 00:03:14.719 formal language, and the outcome. They negotiated 00:03:14.719 --> 00:03:19.280 a settlement down to $33 ,000 from almost $200 00:03:19.280 --> 00:03:22.259 ,000. It just shows how AI can democratize access 00:03:22.259 --> 00:03:25.120 to this highly specialized knowledge, auditing 00:03:25.120 --> 00:03:27.120 knowledge, essentially. It takes some power away 00:03:27.120 --> 00:03:29.680 from these big institutions that rely on information 00:03:29.680 --> 00:03:32.060 asymmetry. That's an incredible result. But does 00:03:32.060 --> 00:03:34.460 this one success story really mean the system 00:03:34.460 --> 00:03:37.099 is fixed? Or does it maybe just highlight how 00:03:37.099 --> 00:03:38.900 broken and complex it is in the first place? 00:03:39.139 --> 00:03:42.199 Designed almost to Well, crush the average person. 00:03:42.340 --> 00:03:44.259 Yeah, that's the question. Complexity is definitely 00:03:44.259 --> 00:03:47.039 the best defense bad actors have right now. But 00:03:47.039 --> 00:03:49.770 AI changes that equation. Pretty rapidly, too. 00:03:49.930 --> 00:03:53.889 OK, so that Claude story really shows the power 00:03:53.889 --> 00:03:56.430 of AI when you give it structured, specialized 00:03:56.430 --> 00:04:00.530 data, those CPT codes. And that idea of specialization 00:04:00.530 --> 00:04:04.770 is key to our next point. Not all AI is an LLM. 00:04:04.810 --> 00:04:07.409 Right. Everyone defaults to that term, LLM, for 00:04:07.409 --> 00:04:09.650 everything. And look, an LLM, a large language 00:04:09.650 --> 00:04:12.150 model, is amazing tech. It's trained specifically 00:04:12.150 --> 00:04:16.100 to predict and generate human -like. text, code, 00:04:16.240 --> 00:04:18.459 that kind of thing. But if LLMs are all you're 00:04:18.459 --> 00:04:20.120 tracking, you're missing the bigger picture. 00:04:20.220 --> 00:04:22.019 There's a whole diversification happening underneath 00:04:22.019 --> 00:04:24.180 the surface. Yeah, our sources were really stressing 00:04:24.180 --> 00:04:26.699 this. You need to know about eight key types 00:04:26.699 --> 00:04:29.220 of AI models to really get where the industry 00:04:29.220 --> 00:04:32.620 is heading in 2025. It's not just about chatbots. 00:04:32.879 --> 00:04:34.800 Absolutely. Think about those specialized agents. 00:04:34.980 --> 00:04:36.360 Everyone's talking about AI that does things 00:04:36.360 --> 00:04:38.279 for you. They often need specialized models. 00:04:38.459 --> 00:04:40.220 LLMs are like the general communicators, right? 00:04:40.300 --> 00:04:42.920 But you need other models for perception or maybe 00:04:42.920 --> 00:04:45.420 organizing information differently. like a specialized 00:04:45.420 --> 00:04:47.980 vision model. Its whole job is image recognition 00:04:47.980 --> 00:04:50.500 and analysis. It doesn't really do text generation 00:04:50.500 --> 00:04:53.100 itself. And that distinction really matters for 00:04:53.100 --> 00:04:55.120 performance. You also have things like embedding 00:04:55.120 --> 00:04:57.860 models. They don't generate text either. Their 00:04:57.860 --> 00:05:00.279 job is to turn abstract concepts, like the meaning 00:05:00.279 --> 00:05:03.480 of a document, into numerical lists, vectors. 00:05:03.620 --> 00:05:07.240 So the AI can understand similarity or meaning, 00:05:07.319 --> 00:05:09.980 not just count words. Exactly. And those generative 00:05:09.980 --> 00:05:11.759 agents, the tools that are starting to perform 00:05:11.759 --> 00:05:15.089 complex tasks. Automatically, they're often specialized 00:05:15.089 --> 00:05:18.110 models running on top of an LLM or alongside 00:05:18.110 --> 00:05:21.509 it. They're tuned for specific actions like researching 00:05:21.509 --> 00:05:24.389 patents, finding sales leads, managing your calendar, 00:05:24.550 --> 00:05:26.949 take perplexity agents, for example. They're 00:05:26.949 --> 00:05:29.430 pitched as these free helpers that can do complex 00:05:29.430 --> 00:05:31.930 research or tasks without you needing to constantly 00:05:31.930 --> 00:05:35.120 guide them step by step. And the usage data kind 00:05:35.120 --> 00:05:37.759 of backs this up, too. We saw a stat, 700 million 00:05:37.759 --> 00:05:40.500 people are using AI now. But the biggest uses 00:05:40.500 --> 00:05:42.740 aren't always the predictable work tasks like 00:05:42.740 --> 00:05:45.360 writing emails. It's often in these more personal, 00:05:45.439 --> 00:05:47.939 maybe complex creative things that need more 00:05:47.939 --> 00:05:50.540 highly tuned, specialized models. Yeah, the really 00:05:50.540 --> 00:05:52.699 sticky utility often hides outside the corporate 00:05:52.699 --> 00:05:55.379 spreadsheet. It makes sense. I have to admit, 00:05:55.500 --> 00:05:57.939 I still wrestle with prompt drift myself sometimes. 00:05:58.060 --> 00:06:00.500 You know, when you're working in a general LLM, 00:06:00.500 --> 00:06:02.740 maybe asking it for creative text and it's... 00:06:02.730 --> 00:06:04.829 It suddenly reverts to that really annoying robotic 00:06:04.829 --> 00:06:09.410 as an AI language model style or overuse of certain 00:06:09.410 --> 00:06:11.970 things like the MDASH. Oh, yeah, the MDASH. That 00:06:11.970 --> 00:06:14.829 just screams AI wrote this, doesn't it? It does. 00:06:14.889 --> 00:06:17.149 And the simple fix I found wasn't just telling 00:06:17.149 --> 00:06:20.129 it don't do that after the fact. It was making 00:06:20.129 --> 00:06:22.949 sure to explicitly include those negative style 00:06:22.949 --> 00:06:26.649 constraints. Don't use MDASHs. Avoid phrases 00:06:26.649 --> 00:06:29.689 like delve into right in the initial prompt or 00:06:29.689 --> 00:06:32.170 the system instructions before it even. starts 00:06:32.170 --> 00:06:34.509 writing. That's a great practical tip. It actually 00:06:34.509 --> 00:06:36.149 highlights why people are moving towards these 00:06:36.149 --> 00:06:38.449 specialized models sometimes to avoid that kind 00:06:38.449 --> 00:06:41.209 of generic output or fighting the model's defaults. 00:06:41.230 --> 00:06:43.310 So, OK, besides just avoiding the jargon, how 00:06:43.310 --> 00:06:45.490 important is it really for the average person, 00:06:45.610 --> 00:06:48.370 you know, someone not building AI to understand 00:06:48.370 --> 00:06:51.189 these different specialized model types? Understanding 00:06:51.189 --> 00:06:53.430 the eight types is key because specialized agents 00:06:53.430 --> 00:06:55.829 will soon perform most complex tasks automatically, 00:06:56.209 --> 00:07:00.019 less fiddly tromping needed. Okay, switching 00:07:00.019 --> 00:07:02.240 gears a bit. Let's talk about the sheer speed 00:07:02.240 --> 00:07:05.019 of things happening now. Product releases, legal 00:07:05.019 --> 00:07:07.970 fights. It's intense. On the creative front, 00:07:08.069 --> 00:07:10.810 big platforms are going all in. Canva, for instance, 00:07:10.949 --> 00:07:14.430 is going almost full AI, they say. Powerful new 00:07:14.430 --> 00:07:16.430 upgrades changing how designers even think about 00:07:16.430 --> 00:07:18.829 templates. And we're seeing these hyper -specialized 00:07:18.829 --> 00:07:22.149 tools pop up constantly. Like Higgs Field AI 00:07:22.149 --> 00:07:24.189 launched something called InstaDump. You give 00:07:24.189 --> 00:07:27.509 it one OK -ish photo and it generates like 15 00:07:27.509 --> 00:07:29.829 different professional, highly stylized shots 00:07:29.829 --> 00:07:32.930 from it in seconds. Pretty wild. Meanwhile, the 00:07:32.930 --> 00:07:34.750 battle for market access getting the next billion 00:07:34.750 --> 00:07:37.459 users onboarded is heating up. Google just gave 00:07:37.459 --> 00:07:40.360 18 months of free AI Pro access to millions in 00:07:40.360 --> 00:07:43.420 India. That mirrors what Perplexity and ChatGPT 00:07:43.420 --> 00:07:45.019 have been doing with free tiers and promotions. 00:07:45.500 --> 00:07:48.040 It's a land grab for future users in key markets. 00:07:48.300 --> 00:07:50.420 Yeah, contrast that sort of market expansion 00:07:50.420 --> 00:07:52.519 with the legal battles brewing, especially here 00:07:52.519 --> 00:07:55.250 in the U .S. There was a ruling. A U .S. judge 00:07:55.250 --> 00:07:57.449 said George R .R. Martin can proceed with his 00:07:57.449 --> 00:08:00.370 lawsuit against OpenAI and Microsoft. The claim 00:08:00.370 --> 00:08:03.910 is ChatGPT basically wrote an unauthorized Game 00:08:03.910 --> 00:08:07.009 of Thrones sequel using his work. OpenAI is definitely 00:08:07.009 --> 00:08:09.769 facing a wave of these copyright and data usage 00:08:09.769 --> 00:08:12.810 lawsuits now. It's that tension, right, between 00:08:12.810 --> 00:08:14.829 the legal headaches of how these things are built 00:08:14.829 --> 00:08:17.089 and the immediate productivity gains they offer. 00:08:17.250 --> 00:08:19.910 Like perplexity just dropped, perplexity patents. 00:08:20.110 --> 00:08:22.110 Makes patent research, which is usually a huge 00:08:22.110 --> 00:08:25.310 pain, super. easy scans blogs papers looking 00:08:25.310 --> 00:08:27.329 for hidden gems traditional search might miss 00:08:27.329 --> 00:08:29.910 and that kind of utility it's out there in free 00:08:29.910 --> 00:08:32.549 beta globally right now and we're seeing a whole 00:08:32.549 --> 00:08:35.549 crop of these like hyper specialized little utility 00:08:35.549 --> 00:08:38.330 agents targeting very specific pain points there's 00:08:38.330 --> 00:08:40.830 get solved .ai it spots text that sounds too 00:08:40.830 --> 00:08:42.649 much like ai and helps make it sound more natural 00:08:42.649 --> 00:08:45.070 chirps an ai agent built just for academic research 00:08:45.070 --> 00:08:47.549 handling citations finding papers then you've 00:08:47.549 --> 00:08:50.149 got free tools like a rest video Does what it 00:08:50.149 --> 00:08:53.190 says. Removes watermarks from AI -generated video, 00:08:53.289 --> 00:08:55.870 maybe from Sora or Google's VO or even TikTok's. 00:08:56.009 --> 00:08:58.090 And in the big enterprise world, the money's 00:08:58.090 --> 00:09:02.169 flowing. Veeam is buying security AI for $1 .7 00:09:02.169 --> 00:09:05.409 billion. That's all about boosting data security 00:09:05.409 --> 00:09:08.429 and giving their, what, 28 ,000 -plus customers 00:09:08.429 --> 00:09:11.350 more control over AI usage. Mid -roll sponsor 00:09:11.350 --> 00:09:13.649 Reed Placeholder provided separately. Okay, now 00:09:13.649 --> 00:09:16.450 we get to the trillion -dollar question. Literally, 00:09:16.610 --> 00:09:19.169 NVIDIA, it's officially the first public company 00:09:19.169 --> 00:09:22.269 ever to be worth $5 trillion. Trillion, with 00:09:22.269 --> 00:09:24.379 a T. Just pause on that number. That valuation 00:09:24.379 --> 00:09:26.679 is bigger than the entire GDP of every single 00:09:26.679 --> 00:09:28.379 country on Earth except for the U .S. and China. 00:09:28.799 --> 00:09:31.220 That level of financial power centralized in 00:09:31.220 --> 00:09:34.100 one hardware company is just astonishing. And 00:09:34.100 --> 00:09:35.840 this isn't just, you know, stock market hype 00:09:35.840 --> 00:09:38.620 detached from reality. The actual demand is ferocious. 00:09:38.700 --> 00:09:41.379 They have already locked in $500 billion, half 00:09:41.379 --> 00:09:43.460 a trillion in actual orders for their AI chips 00:09:43.460 --> 00:09:46.120 just through next year. Whoa. I mean, just imagine 00:09:46.120 --> 00:09:48.519 trying to scale services to handle a billion 00:09:48.519 --> 00:09:50.940 queries. You need that kind of hardware foundation. 00:09:51.080 --> 00:09:54.059 That's why their lead feels so solid right now. 00:09:54.279 --> 00:09:56.860 So what's really fueling this massive valuation, 00:09:57.259 --> 00:09:58.740 it really comes down to the chips themselves, 00:09:58.919 --> 00:10:01.100 right? Their Blackwell chips are seen as the 00:10:01.100 --> 00:10:04.480 undisputed gold standard for the kind of massive 00:10:04.480 --> 00:10:07.139 parallel computing that AI training and inference 00:10:07.139 --> 00:10:09.220 requires. They've already shipped something like 00:10:09.220 --> 00:10:12.720 6 million units, and there are 14 million more 00:10:12.720 --> 00:10:14.899 reportedly coming down the pipeline. Yeah, and 00:10:14.899 --> 00:10:17.210 this is where you see the long game. They aren't 00:10:17.210 --> 00:10:19.409 just selling chips like commodities. They're 00:10:19.409 --> 00:10:21.370 embedding themselves in foundational strategic 00:10:21.370 --> 00:10:24.009 infrastructure. This work is really interesting. 00:10:24.149 --> 00:10:26.789 OK, get this. They have a $1 billion deal with 00:10:26.789 --> 00:10:29.289 Nokia, not for phones, but for building next 00:10:29.289 --> 00:10:31.549 -gen telecom gear. We're talking 5G build -out, 00:10:31.629 --> 00:10:35.149 planning for 6G, using NVIDIA tech. Wow. OK. 00:10:35.250 --> 00:10:37.570 And then there's that huge project with Oracle, 00:10:37.769 --> 00:10:41.870 a multi -year deal to build seven massive supercomputers, 00:10:41.870 --> 00:10:43.889 mostly for U .S. government and research use. 00:10:44.539 --> 00:10:47.320 One of those supercomputers, just one, will contain 00:10:47.320 --> 00:10:51.240 100 ,000 Blackwell chips. Their hardware is literally 00:10:51.240 --> 00:10:54.559 becoming a strategic national asset. And you 00:10:54.559 --> 00:10:56.240 can't ignore the political context here either 00:10:56.240 --> 00:10:57.960 because it shapes their potential market. Remember, 00:10:58.120 --> 00:11:00.899 they used to dominate in China, like 95 percent 00:11:00.899 --> 00:11:03.580 market share for AI chips. Then the U .S. export 00:11:03.580 --> 00:11:05.620 bans hit and that dropped pretty much to zero. 00:11:05.700 --> 00:11:08.700 Huge revenue hit. But now there's been recent 00:11:08.700 --> 00:11:11.639 talk, high level meetings, maybe a political 00:11:11.639 --> 00:11:14.070 reopening of that door. If that happens, it could 00:11:14.070 --> 00:11:17.009 unlock another massive wave of demand and revenue 00:11:17.009 --> 00:11:18.929 for them. Yeah. Even getting hints of political 00:11:18.929 --> 00:11:20.850 favor, like Trump calling the Blackwell chip 00:11:20.850 --> 00:11:23.110 super duper, it all feeds the narrative. And 00:11:23.110 --> 00:11:25.330 the scale of their dominance becomes really clear 00:11:25.330 --> 00:11:27.330 when you compare them. Their current five trillion 00:11:27.330 --> 00:11:30.009 dollar value exceeds the combined market value 00:11:30.009 --> 00:11:33.509 of Intel, AMD, Qualcomm, Broadcom, ARM, ASML 00:11:33.509 --> 00:11:36.379 and TSMC. All of them put together. That single 00:11:36.379 --> 00:11:38.320 comparison just shows you where the critical 00:11:38.320 --> 00:11:40.919 bottleneck and therefore the profit center of 00:11:40.919 --> 00:11:43.500 this whole AI revolution sits right now. It's 00:11:43.500 --> 00:11:46.340 the hardware. So does that create a problem? 00:11:46.519 --> 00:11:49.759 Does having such intense financial and technological 00:11:49.759 --> 00:11:52.399 centralization around basically one hardware 00:11:52.399 --> 00:11:54.720 provider, does that create a critical point of 00:11:54.720 --> 00:11:57.710 failure for the entire AI industry? Yes. Definitely. 00:11:58.190 --> 00:12:00.129 Reliance on one chip standard absolutely risks 00:12:00.129 --> 00:12:02.710 vendor lock -in, future bottlenecks, price heights. 00:12:03.309 --> 00:12:06.090 But their ecosystem, the software, the tools 00:12:06.090 --> 00:12:08.509 built around their chips, it's currently unmatched. 00:12:08.590 --> 00:12:10.090 Makes it a really hard risk for companies to 00:12:10.090 --> 00:12:13.590 avoid right now. So let's recap the big idea 00:12:13.590 --> 00:12:16.330 from today. We saw AI moving beyond just being 00:12:16.330 --> 00:12:19.009 a tool for office productivity. It's becoming 00:12:19.009 --> 00:12:21.009 a tool for justice in a way, taking care of these 00:12:21.009 --> 00:12:23.269 really complex, confusing systems that are often 00:12:23.269 --> 00:12:25.330 designed, frankly, to extract money from ordinary 00:12:25.330 --> 00:12:27.970 people. At the same time, the underlying capability 00:12:27.970 --> 00:12:31.309 of AI is diversifying like crazy. We talked about 00:12:31.309 --> 00:12:33.750 the eight model types. It's way beyond just LLMs 00:12:33.750 --> 00:12:35.950 now. But the foundational infrastructure that 00:12:35.950 --> 00:12:38.610 runs all this amazing stuff, the hardware, is 00:12:38.610 --> 00:12:40.470 doing the opposite. It's centralizing incredibly 00:12:40.470 --> 00:12:43.299 around NVIDIA. And that's creating these huge 00:12:43.299 --> 00:12:45.419 strategic dependencies, even at a national level. 00:12:45.559 --> 00:12:47.419 So what does this all mean for you? Well, think 00:12:47.419 --> 00:12:50.580 about it. If a powerful AI like Claude fed the 00:12:50.580 --> 00:12:52.980 right data can successfully challenge a nearly 00:12:52.980 --> 00:12:56.519 $200 ,000 hospital bill, what overly complex 00:12:56.519 --> 00:12:58.840 system, what bureaucracy, what confusing contract 00:12:58.840 --> 00:13:01.399 in your own life is just waiting for a simple, 00:13:01.539 --> 00:13:04.440 well -phrased, data -rich prompt? The knowledge, 00:13:04.559 --> 00:13:06.379 the tools to fight back against that complexity, 00:13:06.559 --> 00:13:08.679 they're becoming accessible like never before. 00:13:08.899 --> 00:13:11.139 The application, though, that's up to you. Thank 00:13:11.139 --> 00:13:12.669 you for... diving deep with us today. We really 00:13:12.669 --> 00:13:13.909 hope you feel thoroughly well informed.